http://community.qnx.com/sf/discussion/do/listTopics/projects.networking/discussion.general Posts for General http://community.qnx.com/sf/go/post101354 Try ftp or scp. On Mon, May 13, 2013 at 01:44:54PM -0400, Reginald White wrote: > Unfortunately, I can't make changes to the existing QNX systems already in the field. > So if a Samba server is needed on each QNX system (~3000), I won't be able to use that solution. > > Do you have anything else I could possible try? > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101353 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-05-13T17:45:30Z http://community.qnx.com/sf/go/post101353 Unfortunately, I can't make changes to the existing QNX systems already in the field. So if a Samba server is needed on each QNX system (~3000), I won't be able to use that solution. Do you have anything else I could possible try? Reginald White 2013-05-13T17:44:54Z http://community.qnx.com/sf/go/post101352 There's a scp / ssh client for windows called 'putty'. samba is the server side that would run on the qnx machines that would allow standard SMB/CIFS access from any windows box (would provide a windows share). You could use the standard UNC path api in windows. There's one here: http://ftp.netbsd.org/pub/pkgsrc/packages/QNX/i386/6.5.0_head_20110826/All/ Or you could build a slightly newer one here: http://community.qnx.com/sf/projects/pkgsrc http://community.qnx.com/sf/wiki/do/viewPage/projects.pkgsrc/wiki/BinaryPackages samba isn't officially supported. On Mon, May 13, 2013 at 01:23:08PM -0400, Reginald White wrote: > I'm not familiar with scp or samba. > > Are these available to Windows so that I can create an application around it (API) to upload file the QNX system and replace existing files within the file directories? > > If so, how and where do I get them? > > Thanks > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101349 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-05-13T17:35:47Z http://community.qnx.com/sf/go/post101349 I'm not familiar with scp or samba. Are these available to Windows so that I can create an application around it (API) to upload file the QNX system and replace existing files within the file directories? If so, how and where do I get them? Thanks Reginald White 2013-05-13T17:23:08Z http://community.qnx.com/sf/go/post101348 On Mon, May 13, 2013 at 01:10:53PM -0400, Reginald White wrote: > What is this product? http://www.qnx.com/developers/docs/6.5.0_sp1/topic/com.qnx.doc.neutrino_sys_arch/qnet.html?cp=13_10_13 > > I'm looking to be able to connect to my systems that are run varies versions of QNX. > Once I connect, I would like to be able upload files to those systems. > It seem like QNet can handle this, but it does seem like I can't access it from a Windows 7 system. Correct. > > Does TDP allow my Windows 7 systems to gain access to the QNet services running on my QNX systems? No. > > If not, where else should I turn. The simplest may be ftp. There's also scp or samba. > > Also, I can't modify these existing QNX systems, so that is why I'm look to QNet for assistants. > > Thanks > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101346 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-05-13T17:16:55Z http://community.qnx.com/sf/go/post101346 What is this product? I'm looking to be able to connect to my systems that are run varies versions of QNX. Once I connect, I would like to be able upload files to those systems. It seem like QNet can handle this, but it does seem like I can't access it from a Windows 7 system. Does TDP allow my Windows 7 systems to gain access to the QNet services running on my QNX systems? If not, where else should I turn. Also, I can't modify these existing QNX systems, so that is why I'm look to QNet for assistants. Thanks Reginald White 2013-05-13T17:10:53Z http://community.qnx.com/sf/go/post101197 Thank you very much. It does work! > > struct ifdatareq ifdr; > memset(&ifdr, 0, sizeof(ifdr)); > strlcpy(ifdr.ifdr_name, ifname, sizeof(ifdr.ifdr_name)); > ioctl(s, SIOCGIFDATA, &ifdr); > > On Sat, May 04, 2013 at 08:52:36AM +0800, Allen Chen wrote: > > Thanks for your reply. Is there any C/C++ codes to implement it? I want > > to send these status data by UPnP. > > > > Allen > > > > > 'ifconfig -v' gives packet counts. > > > > > > You'll have to elaborate if you want something more. > > > > > > On Fri, May 03, 2013 at 11:09:33PM +0800, Allen Chen wrote: > > >> lt means network interface card flow. l want to monitor system status. > > >> > > >> Allen > > >> > > >>> What exactly do you mean by 'NIC flow'? > > >>> > > >>> On Thu, May 02, 2013 at 11:48:45PM -0400, Allen Chen wrote: > > >>>> On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX > Neutrino 6.5.0, there is no such files. How to do it? > > >>>> > > >>>> Allen Chen > > >>>> > > >>>> > > >>>> > > >>>> _______________________________________________ > > >>>> > > >>>> General > > >>>> http://community.qnx.com/sf/go/post101106 > > >>>> To cancel your subscription to this discussion, please e-mail general- > networking-unsubscribe@community.qnx.com > > >>> > > >>> > > >>> > > >>> _______________________________________________ > > >>> > > >>> General > > >>> http://community.qnx.com/sf/go/post101113 > > >>> To cancel your subscription to this discussion, please e-mail general- > networking-unsubscribe@community.qnx.com > > >> > > >> > > >> > > >> _______________________________________________ > > >> > > >> General > > >> http://community.qnx.com/sf/go/post101122 > > >> To cancel your subscription to this discussion, please e-mail general- > networking-unsubscribe@community.qnx.com > > > > > > > > > > > > _______________________________________________ > > > > > > General > > > http://community.qnx.com/sf/go/post101123 > > > To cancel your subscription to this discussion, please e-mail general- > networking-unsubscribe@community.qnx.com > > > > > > > > > > > > > > > _______________________________________________ > > > > General > > http://community.qnx.com/sf/go/post101147 > > To cancel your subscription to this discussion, please e-mail general- > networking-unsubscribe@community.qnx.com Allen Chen 2013-05-07T06:53:00Z http://community.qnx.com/sf/go/post101178 struct ifdatareq ifdr; memset(&ifdr, 0, sizeof(ifdr)); strlcpy(ifdr.ifdr_name, ifname, sizeof(ifdr.ifdr_name)); ioctl(s, SIOCGIFDATA, &ifdr); On Sat, May 04, 2013 at 08:52:36AM +0800, Allen Chen wrote: > Thanks for your reply. Is there any C/C++ codes to implement it? I want > to send these status data by UPnP. > > Allen > > > 'ifconfig -v' gives packet counts. > > > > You'll have to elaborate if you want something more. > > > > On Fri, May 03, 2013 at 11:09:33PM +0800, Allen Chen wrote: > >> lt means network interface card flow. l want to monitor system status. > >> > >> Allen > >> > >>> What exactly do you mean by 'NIC flow'? > >>> > >>> On Thu, May 02, 2013 at 11:48:45PM -0400, Allen Chen wrote: > >>>> On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX Neutrino 6.5.0, there is no such files. How to do it? > >>>> > >>>> Allen Chen > >>>> > >>>> > >>>> > >>>> _______________________________________________ > >>>> > >>>> General > >>>> http://community.qnx.com/sf/go/post101106 > >>>> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > >>> > >>> > >>> > >>> _______________________________________________ > >>> > >>> General > >>> http://community.qnx.com/sf/go/post101113 > >>> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > >> > >> > >> > >> _______________________________________________ > >> > >> General > >> http://community.qnx.com/sf/go/post101122 > >> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > > > > > > > > _______________________________________________ > > > > General > > http://community.qnx.com/sf/go/post101123 > > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > > > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101147 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-05-06T14:33:51Z http://community.qnx.com/sf/go/post101147 Thanks for your reply. Is there any C/C++ codes to implement it? I want to send these status data by UPnP. Allen > 'ifconfig -v' gives packet counts. > > You'll have to elaborate if you want something more. > > On Fri, May 03, 2013 at 11:09:33PM +0800, Allen Chen wrote: >> lt means network interface card flow. l want to monitor system status. >> >> Allen >> >>> What exactly do you mean by 'NIC flow'? >>> >>> On Thu, May 02, 2013 at 11:48:45PM -0400, Allen Chen wrote: >>>> On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX Neutrino 6.5.0, there is no such files. How to do it? >>>> >>>> Allen Chen >>>> >>>> >>>> >>>> _______________________________________________ >>>> >>>> General >>>> http://community.qnx.com/sf/go/post101106 >>>> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com >>> >>> >>> >>> _______________________________________________ >>> >>> General >>> http://community.qnx.com/sf/go/post101113 >>> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com >> >> >> >> _______________________________________________ >> >> General >> http://community.qnx.com/sf/go/post101122 >> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101123 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > Allen Chen 2013-05-04T00:51:57Z http://community.qnx.com/sf/go/post101123 'ifconfig -v' gives packet counts. You'll have to elaborate if you want something more. On Fri, May 03, 2013 at 11:09:33PM +0800, Allen Chen wrote: > lt means network interface card flow. l want to monitor system status. > > Allen > > > > > What exactly do you mean by 'NIC flow'? > > > > On Thu, May 02, 2013 at 11:48:45PM -0400, Allen Chen wrote: > >> On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX Neutrino 6.5.0, there is no such files. How to do it? > >> > >> Allen Chen > >> > >> > >> > >> _______________________________________________ > >> > >> General > >> http://community.qnx.com/sf/go/post101106 > >> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > > > > > > > > > > _______________________________________________ > > > > General > > http://community.qnx.com/sf/go/post101113 > > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101122 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-05-03T15:13:58Z http://community.qnx.com/sf/go/post101122 lt means network interface card flow. l want to monitor system status. Allen > > What exactly do you mean by 'NIC flow'? > > On Thu, May 02, 2013 at 11:48:45PM -0400, Allen Chen wrote: >> On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX Neutrino 6.5.0, there is no such files. How to do it? >> >> Allen Chen >> >> >> >> _______________________________________________ >> >> General >> http://community.qnx.com/sf/go/post101106 >> To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101113 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Allen Chen 2013-05-03T15:09:50Z http://community.qnx.com/sf/go/post101113 What exactly do you mean by 'NIC flow'? On Thu, May 02, 2013 at 11:48:45PM -0400, Allen Chen wrote: > On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX Neutrino 6.5.0, there is no such files. How to do it? > > Allen Chen > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post101106 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-05-03T13:22:18Z http://community.qnx.com/sf/go/post101106 On Linux, I can get NIC flow extracted from /proc/net/dev. But on QNX Neutrino 6.5.0, there is no such files. How to do it? Allen Chen Allen Chen 2013-05-03T03:48:45Z http://community.qnx.com/sf/go/post101008 If you run 'use /lib/dll/devn-smsc9500.so' you should see the correct command line options. Yes, the documentation should be corrected. On 2013-04-30 10:17 AM, "Oleg V" <community-noreply@qnx.com> wrote: >That worked, thanks a lot! I guess the documentation is not quite >complete or correct. It states busno/devno: > >http://www.qnx.org.uk/developers/docs/6.5.0/index.jsp?topic=%2Fcom.qnx.doc >.neutrino_utilities%2Fi%2Fio-pkt.html > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post101007 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-04-30T14:19:48Z http://community.qnx.com/sf/go/post101007 That worked, thanks a lot! I guess the documentation is not quite complete or correct. It states busno/devno: http://www.qnx.org.uk/developers/docs/6.5.0/index.jsp?topic=%2Fcom.qnx.doc.neutrino_utilities%2Fi%2Fio-pkt.html Oleg V 2013-04-30T14:17:38Z http://community.qnx.com/sf/go/post101001 OK, so try busnum=x,devnum=y. On 2013-04-30 9:50 AM, "Oleg V" <community-noreply@qnx.com> wrote: > >Oh I completely forgot about system log. Thanks Hugh! > >It seems smsc9500 driver doesn't quite like busno parameter for whatever >reason. > >Jan 01 00:03:06 3 14 0 devn-smsc9500: Unknown option busno=1 >Jan 01 00:03:06 2 14 0 shim: unable to init dll >devn-smsc9500.so: Invalid argument >Jan 01 00:03:06 2 14 0 Unable to init devn-smsc9500.so: Unknown >error > > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post101000 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-04-30T13:57:54Z http://community.qnx.com/sf/go/post101000 Oh I completely forgot about system log. Thanks Hugh! It seems smsc9500 driver doesn't quite like busno parameter for whatever reason. Jan 01 00:03:06 3 14 0 devn-smsc9500: Unknown option busno=1 Jan 01 00:03:06 2 14 0 shim: unable to init dll devn-smsc9500.so: Invalid argument Jan 01 00:03:06 2 14 0 Unable to init devn-smsc9500.so: Unknown error Oleg V 2013-04-30T13:50:58Z http://community.qnx.com/sf/go/post100997 Are there any messages in sloginfo? On 2013-04-30 8:48 AM, "Oleg V" <community-noreply@qnx.com> wrote: >Nope, tried that already. Getting the same "unknown error". I think it >just generally doesn't like the vid/did/busno/devno parameters for >whatever reason. > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post100995 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-04-30T13:11:28Z http://community.qnx.com/sf/go/post100995 Nope, tried that already. Getting the same "unknown error". I think it just generally doesn't like the vid/did/busno/devno parameters for whatever reason. Oleg V 2013-04-30T12:48:12Z http://community.qnx.com/sf/go/post100994 Does it work if you run io-pkt as follows: io-pkt-v4-hc -dsmsc9500 busno=x,devno=y,mac=001122334455 -dsmsc9500 busno=a,devno=b,mac=001122334456 On 2013-04-29 5:16 PM, "Oleg V" <community-noreply@qnx.com> wrote: > >The problem I'm facing is with two smsc9500 USB Ethernet devices. In >fact, multiple instances config seems to be supported just fine. The real >problem is with assigning MAC address to individual instances. > >The following would recognize and crate two interfaces en0 and en1 just >as expected, but MAC addresses will be 00-00-00-00-00-00 for both: > >io-pkt-v4-hc -vvvv -ptcpip -dsmsc9500 > >The following would recognize and assign the same MAC for both interfaces >en0 and en1: > >io-pkt-v4-hc -vvvv -ptcpip -dsmsc9500 mac=001122334455 > >Any combination with using vid/did/busno/devno would simply result into >"devn-smsc9500.so: unknown error" error. Either I'm missing something or >there is a bug in the deriver. For example the following: > >io-pkt-v4-hc -vvvv -ptcpip >-dsmsc9500,vid=0x0424,did=0xec00,busno=1,devno=2,mac=001122334455 > >would generate the same "unknown error". Obviously busno/devno parameters >are correct as they are copy/pasted from "usb -v" output. > >So my idea with using busno/devno to assign per device MAC address for >two smsc9500 chips appears to be useless. > >Does anyone have an idea on why it doesn't like vid/did/busno/devno >setup? Alternatively, any other ways to assign different MACs so I can >use both interfaces? > >Thanks for any pointers. > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post100980 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-04-30T12:26:56Z http://community.qnx.com/sf/go/post100981 Sorry, a small typo in that last command line: io-pkt-v4-hc -vvvv -ptcpip -dsmsc9500 vid=0x0424,did=0xec00,busno=1,devno=2,mac=001122334455 The above results into "unknown error" even though vid/did/busno/devno are all correct... Oleg V 2013-04-29T21:25:48Z http://community.qnx.com/sf/go/post100980 The problem I'm facing is with two smsc9500 USB Ethernet devices. In fact, multiple instances config seems to be supported just fine. The real problem is with assigning MAC address to individual instances. The following would recognize and crate two interfaces en0 and en1 just as expected, but MAC addresses will be 00-00-00-00-00-00 for both: io-pkt-v4-hc -vvvv -ptcpip -dsmsc9500 The following would recognize and assign the same MAC for both interfaces en0 and en1: io-pkt-v4-hc -vvvv -ptcpip -dsmsc9500 mac=001122334455 Any combination with using vid/did/busno/devno would simply result into "devn-smsc9500.so: unknown error" error. Either I'm missing something or there is a bug in the deriver. For example the following: io-pkt-v4-hc -vvvv -ptcpip -dsmsc9500,vid=0x0424,did=0xec00,busno=1,devno=2,mac=001122334455 would generate the same "unknown error". Obviously busno/devno parameters are correct as they are copy/pasted from "usb -v" output. So my idea with using busno/devno to assign per device MAC address for two smsc9500 chips appears to be useless. Does anyone have an idea on why it doesn't like vid/did/busno/devno setup? Alternatively, any other ways to assign different MACs so I can use both interfaces? Thanks for any pointers. Oleg V 2013-04-29T21:16:04Z http://community.qnx.com/sf/go/post100807 > > It works since ~15 years :) Yes, if you see a connection like this on your samba-server, where qnx is the client: $ netstat -an | grep 445 tcp 0 0 0.0.0.0:445 0.0.0.0:* LISTEN tcp 0 0 10.0.0.77:445 10.0.12.10:49333 ESTABLISHED Well, I bet you will see something like this: $ netstat -an | grep 139 tcp 0 0 0.0.0.0:139 0.0.0.0:* LISTEN tcp 0 0 10.0.0.77:139 10.0.12.10:1044 ESTABLISHED -Michael > > --Armin > > Michael Tasche wrote: > > Hi, > > > > no answer meens port 445 is not supported? > > > > -Michael > > > > > > > > > > > > > > > > _______________________________________________ > > > > General > > http://community.qnx.com/sf/go/post100789 > > To cancel your subscription to this discussion, please e-mail general- > networking-unsubscribe@community.qnx.com > > > Michael Tasche 2013-04-23T14:03:57Z http://community.qnx.com/sf/go/post100799 It works since ~15 years :) --Armin Michael Tasche wrote: > Hi, > > no answer meens port 445 is not supported? > > -Michael > > > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post100789 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > Armin Steinhoff 2013-04-23T13:07:14Z http://community.qnx.com/sf/go/post100789 Hi, no answer meens port 445 is not supported? -Michael Michael Tasche 2013-04-23T08:53:08Z http://community.qnx.com/sf/go/post100787 Hi, A prospect is asking if it is possible to run qnet across CAN? Thanks, Garry Garry Bleasdale 2013-04-23T08:40:16Z http://community.qnx.com/sf/go/post100692 IT WORK! Thanks for your good answer!! Gervot Olivier 2013-04-17T14:18:46Z http://community.qnx.com/sf/go/post100682 Try: signal(SIGPIPE, SIG_IGN); Sent from my BlackBerry 10 smartphone on the Rogers network. From: Gervot Olivier Sent: Wednesday, April 17, 2013 8:28 AM To: general-networking Reply To: general-networking@community.qnx.com Subject: Socket deconnection Hi I have to manage a very busy socket in a client connection I detect disconnection by testing the return value of the send() function int size_written = send( fd, .... , ....); then i test the errno variable if (errno == ECONNRESET || errno == EPIPE) { fd = CODE_SOCKET_CLOSE; return CODE_DECONNEXION; } The socket is used both in a thread for reading and 2 timers for writing It work but when the socket trafic is increased and a server deconnection occured, for an unknown reason the the process is stopped. There is no error and no core dump. It's like an exit function call Thank for your answer _______________________________________________ General http://community.qnx.com/sf/go/post100676 To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-04-17T13:20:45Z http://community.qnx.com/sf/go/post100676 Hi I have to manage a very busy socket in a client connection I detect disconnection by testing the return value of the send() function int size_written = send( fd, .... , ....); then i test the errno variable if (errno == ECONNRESET || errno == EPIPE) { fd = CODE_SOCKET_CLOSE; return CODE_DECONNEXION; } The socket is used both in a thread for reading and 2 timers for writing It work but when the socket trafic is increased and a server deconnection occured, for an unknown reason the the process is stopped. There is no error and no core dump. It's like an exit function call Thank for your answer Gervot Olivier 2013-04-17T12:28:03Z http://community.qnx.com/sf/go/post100629 Did some checks it seems the messages get dropped. # netstat -p udp udp: 4806 datagrams received 0 with incomplete header 0 with bad data length field 0 with bad checksum 74 dropped due to no socket 4725 broadcast/multicast datagrams dropped due to no socket 0 dropped due to full socket buffers 7 delivered 81 PCB hash misses 3896 datagrams output Bernd Huegel 2013-04-16T12:40:52Z http://community.qnx.com/sf/go/post100628 Hi, i am trying to receive broadcastmessages that i send to the IP address 10.1.1.255. All my target are in this subnet. The problem is that all broadcastmessages seem to be ignored by the qnx targets. The targets plattforms in my subnet are a mix of Win7, WinXp, WinCE and QNX6.5. And with the exception of the qnx target it works fine. Unicast to the qnx targets work. I use the function recvFrom and setsockopt with SO_BROADCAST as parameter after opening the socket. What else do i have to do to receive the broadcasts? Any help would be a appreciated. If a more detailed description is needed plz tell me :) Bernd Huegel 2013-04-16T11:18:56Z http://community.qnx.com/sf/go/post100424 I have fixed it. __QNX__ should be added to the source codes to support QNX platform. Allen Chen 2013-04-10T07:46:23Z http://community.qnx.com/sf/go/post100401 Hi all, does fs-cifs support SMB over TCPIP (Normally port 445) ? I only get connctions using NETBIOS over TCPIP ( Normally port 139). Kind Regards Michael Michael Tasche 2013-04-09T12:52:01Z http://community.qnx.com/sf/go/post100364 On QNX Neutrino 6.5.0, libupnp-1.6.17 cannot be compiled smoothly. But the sample program tv_device and tv_ctrlpt cannot run simultaneously. In one terminal, run tv_device #./tv_device -ip 192.168.137.128 -port 49152 ... Ok I another terminal, run tv_ctrlpt (192.168.137.128:49152) #./tv_ctrlpt UpnpInit() Error -203 -203 is coded as UPNP_E_SOCKET_BIND. I have configured the Neutrino with lo0 and en0 interfaces. On other OS, for example Ubuntu, there is no such problem. What is the reason? Allen Chen 2013-04-08T07:57:00Z http://community.qnx.com/sf/go/post100078 That works..! Thanks a GIG!! Rajsree Vijayan 2013-03-21T21:24:57Z http://community.qnx.com/sf/go/post100070 <font face="Default Sans Serif,Verdana,Arial,Helvetica,sans-serif" size="2">Use the route command to set the default gateway to be the router. Then, forward port 80 on the DSL router to the IP of your QNX box.<br><br><font color="#990099">-----Rajsree Vijayan <community-noreply@qnx.com> wrote: -----</font><div style="padding-left:5px;"><div style="padding-right:0px;padding-left:5px;border-left:solid black 2px;">To: general-networking <general-networking@community.qnx.com><br>From: Rajsree Vijayan <community-noreply@qnx.com><br>Date: 03/21/2013 09:32AM<br>Subject: Re: SSH and Slinger (Web) through a DSL router<br><br><div><font face="Courier New,Courier,monospace" size="3">I face the same problem with QNX 6.3. Do you have a solution for it yet?<br><br><br><br>_______________________________________________<br><br>General<br><a href="http://community.qnx.com/sf/go/post100069">http://community.qnx.com/sf/go/post100069</a><br>To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com<br></font></div></div></div><div></div></font> Todd Peterson 2013-03-21T16:29:58Z http://community.qnx.com/sf/go/post100069 I face the same problem with QNX 6.3. Do you have a solution for it yet? Rajsree Vijayan 2013-03-21T14:49:37Z http://community.qnx.com/sf/go/post99945 rval= socket(AF_INET, SOCK_RAW,6); I am trying to make socket by using above function but it is showing error " undefined reference to Socket" Please tell me how to link against lib socket in detail ,i am not able to link it . Thanks pooja chahal 2013-03-16T03:49:59Z http://community.qnx.com/sf/go/post99784 You would need to set it prior to starting your app: SOCK=/net1 youAppGoesHere This means that this instance of your application will use the SOCK=/net1 environment variable. You can not change it once it is set as far as I know, so you would need to launch another instance with SOCK=/net2 to connect to the other stack. Gervais Mulongoy 2013-03-11T12:27:25Z http://community.qnx.com/sf/go/post99758 In the old thread, it sounded like an enhanced or fixed tftpd would be in 6.4.1 . Today I tried putting a 6.5.0 tftpd onto the 6.3.2 system, and tftp-ing locally, I couldn't get any data transfer. That is, on the 6.3.2 system (server), I did "tftp 127.0.0.1", and some setup commands "ver", "tra", & "bin", and then get nnnnnnnnnnnn get /nnnnnnnnnnnn get /tftpboot/nnnnnnnnnnnn but I couldn't get any data transfer, so maybe I was naive to expect a binary from /usr/sbin to interoperate between the 6.X os-es? Anyway, I'm still blocked. :( Keep pursuing pxqldr? Or give up on pxqldr and start trying to ramp up on grub-2.00 ? Thanks, John R. John Ruckstuhl 2013-03-08T16:47:28Z http://community.qnx.com/sf/go/post99750 Great, thanks!!! However, how can I define SOCK=/net1 when I am working in a C++ environment? For example, if I want to use the function getifaddrs(), how to get interfaces linked to SOCK=/dev/net1? Many thanks -----Message d'origine----- From: Gervais Mulongoy Sent: Monday, March 04, 2013 2:14 PM To: general-networking@community.qnx.com Subject: Re: Raw ethernet packet using bpf0 and bpf1 You will need to start netmanager manually for each instance of io-pkt. In my previous example, you would do something like this: SOCK=/net1 netmanager SOCK=/net2 netmanager _______________________________________________ General http://community.qnx.com/sf/go/post99662 To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Michele Damoiseau 2013-03-08T13:07:53Z http://community.qnx.com/sf/go/post99747 i want know qnx 4.5.0 suppots FTP on SSL. i couldnt find out about it. whenever i searched i ended up with SFTP on web. if it is supported can you provide me with a link to download the utility kannan kumar 2013-03-08T08:53:56Z http://community.qnx.com/sf/go/post99743 Four years ago, Sean Boudreau writes that there's a patched tftpd: They're now in the public networking repo: http://community.qnx.com/svn/repos/core_networking/trunk/services/tftpd http://community.qnx.com/svn/repos/core_networking/trunk/utils/t/tftp -seanb but I cannot access those locations; maybe they are no longer available or moved? I'm having the same trouble as described by OP (Original Poster). The server is QNX 6.3.2 . I can see that my tftpd won't honor the request to get "nnnnnnnnnnnn", but it will respond to get "/tftpboot/nnnnnnnnnnnn" . This, btw, is consistent with some QNX docs saying that tftpd expects the filename to begin with "/". OP also responded that finally he tweaked his inetd.conf entry to tftp dgram udp wait root /usr/sbin/tftpd in.tftpd /tftpboot and that worked for with his original tftpd... allowing his pxqldr to pull "nnnnnnnnnnnn". but, that does NOT work for me and this QNX 6.3.2 tftpd So... is there a tftpd (pre-built) that I should be downloading & using? Should I give up on pxqldr (which is in the "bazaar" and has "Maturity" = "Prototype"? Other recent QNX docs suggest downloading a GRUB pre-built executable, but I'm finding source tarballs for 1.99 & 2.00, but no pre-built executables. Thanks for any comments! John R. John Ruckstuhl 2013-03-07T22:12:56Z http://community.qnx.com/sf/go/post99662 You will need to start netmanager manually for each instance of io-pkt. In my previous example, you would do something like this: SOCK=/net1 netmanager SOCK=/net2 netmanager Gervais Mulongoy 2013-03-04T13:15:00Z http://community.qnx.com/sf/go/post99660 Actually, I was launching io-pkt via enum utilities. I have tried after that manually, but my mistake was not to define pci index of each interface... Thanks for that information. Now, I could have now /net1/dev/bpf0 and /net2/dev/bpf0! But a last question, how can I launch netmanager after that, in order to define IP address on each interface? How to pass the prefix to all the network applications? Many thanks Regards Michele -----Message d'origine----- From: Gervais Mulongoy Sent: Friday, March 01, 2013 2:39 PM To: general-networking@community.qnx.com Subject: Re: Raw ethernet packet using bpf0 and bpf1 I just tried the following: 1) io-pkt-v4 -dpcnet -i1 -ptcpip prefix=/net1 2) io-pkt-v4 -de1000 -i2 -ptcpip prefix=/net2 Each instance of Io-pkt lives at its own prefix. Note that this means that all network apps will need to be told which prefix to use or they will complain that they can't find the network stack at the default prefix of /. Further note, you can configure one instance to live at the default location, just omit the "-ptcpip prefx=.*" bit. Following the above steps, I now have two distinct sets of bpf interfaces. /net1/dev/bpf* /net2/dev/bpf* If I want to use the first bpf interface, I specify the prefix: SOCK=/net1 myNetApp If I want to use the second bpf interface, I specify the prefix: SOCK=/net2 myNetApp How are you launching both instances of io-pkt? _______________________________________________ General http://community.qnx.com/sf/go/post99629 To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Michele Damoiseau 2013-03-04T10:38:51Z http://community.qnx.com/sf/go/post99629 I just tried the following: 1) io-pkt-v4 -dpcnet -i1 -ptcpip prefix=/net1 2) io-pkt-v4 -de1000 -i2 -ptcpip prefix=/net2 Each instance of Io-pkt lives at its own prefix. Note that this means that all network apps will need to be told which prefix to use or they will complain that they can't find the network stack at the default prefix of /. Further note, you can configure one instance to live at the default location, just omit the "-ptcpip prefx=.*" bit. Following the above steps, I now have two distinct sets of bpf interfaces. /net1/dev/bpf* /net2/dev/bpf* If I want to use the first bpf interface, I specify the prefix: SOCK=/net1 myNetApp If I want to use the second bpf interface, I specify the prefix: SOCK=/net2 myNetApp How are you launching both instances of io-pkt? Gervais Mulongoy 2013-03-01T13:39:39Z http://community.qnx.com/sf/go/post99627 Wait you launched both and it only created a single bpf dev entry? I'll have to look at the implementation to confirm. On 02/28/2013 12:02 PM, Michele Damoiseau wrote: > Do you mean launch io-pkt for each network interface? > This is already done... > > > -----Message d'origine----- > From: Gervais Mulongoy > Sent: Thursday, February 28, 2013 2:50 PM > To: general-networking@community.qnx.com > Subject: Re: Raw ethernet packet using bpf0 and bpf1 > > As far as I know you will need two instances of io-pkt in order to > achieve that. > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post99607 > To cancel your subscription to this discussion, please e-mail > general-networking-unsubscribe@community.qnx.com > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post99611 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Gervais Mulongoy 2013-03-01T13:29:00Z http://community.qnx.com/sf/go/post99628 Wait you launched both and it only created a single bpf dev entry? I'll have to look at the implementation to confirm. On 02/28/2013 12:02 PM, Michele Damoiseau wrote: > Do you mean launch io-pkt for each network interface? > This is already done... > > > -----Message d'origine----- > From: Gervais Mulongoy > Sent: Thursday, February 28, 2013 2:50 PM > To: general-networking@community.qnx.com > Subject: Re: Raw ethernet packet using bpf0 and bpf1 > > As far as I know you will need two instances of io-pkt in order to > achieve that. > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post99607 > To cancel your subscription to this discussion, please e-mail > general-networking-unsubscribe@community.qnx.com > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post99611 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Gervais Mulongoy 2013-03-01T13:28:49Z http://community.qnx.com/sf/go/post99611 Do you mean launch io-pkt for each network interface? This is already done... -----Message d'origine----- From: Gervais Mulongoy Sent: Thursday, February 28, 2013 2:50 PM To: general-networking@community.qnx.com Subject: Re: Raw ethernet packet using bpf0 and bpf1 As far as I know you will need two instances of io-pkt in order to achieve that. _______________________________________________ General http://community.qnx.com/sf/go/post99607 To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Michele Damoiseau 2013-02-28T17:02:57Z http://community.qnx.com/sf/go/post99607 As far as I know you will need two instances of io-pkt in order to achieve that. Gervais Mulongoy 2013-02-28T13:50:34Z http://community.qnx.com/sf/go/post99508 Hi, I am using QNX 6.5 Is it possible to have more than one bpf created. I would like to have two bpf one named /dev/bpf0 /dev/bpf1 So i can use those two devices bound to the two network adapters of my motherboard. Thank you Michele Damoiseau 2013-02-25T15:15:34Z http://community.qnx.com/sf/go/post99500 <p>When you are looking for a good alternative to high fashion, replica products can definitely be an option. Of course, not all replica products are high quality so you should keep your eyes open when you buy. The <a href="http://8watches.com/replica-tag-heuer-monaco.html">Tag Heuer Monaco cool watches</a> original brand is often too expensive for the average person and many people find replica products more satisfying when it comes to the price. The <a href="http://www.replicasshop.com/bell-ross/military-watch.html">Bell & Ross Military Watches</a> same thing applies for replica watches without having to worry about their quality. Replica watches have <a href="http://www.carewatches.com/breguet-replica/heritage-watch.html">Breguet Heritage Replica Watches</a> the best quality while they look and feel exactly like the original products.</p> <p>Some people buy replica watches online, which is a fast <a href="http://www.watchesgg.com/tissot-t-tracx/">Automatic Tissot T-Tracx Watches</a> and easy way to buy watches as long as you make sure that the company is legitimate. Watches are one of the best gifts and they are always suitable for any occasion; it doesn't matter if you buy one for your wife, your girlfriend or your dad, it will be one of the best gifts they had.</p> Jack Che 2013-02-25T09:23:00Z http://community.qnx.com/sf/go/post99397 Hi, I am running the devc-serpci driver with a Kontron card (CP346) vid=0x1415, did=0x9501 which is presumably compatible with this driver. When reading there is no problem but when I am sending data the only way to get the data trough is to write 16 bytes at the time and flush the output. As you can imagine this is slower than sucking melasa through a straw. Can anyone give me some hints I found this problem was reported before but none one addresses it. (http://community.qnx.com/sf/discussion/do/listPosts/projects.bsp/discussion.bsp.topc7626). Running QNX version 6.5. Thank you, Javier Settings: handle_ = ::open(config_.modem_device_.c_str(), O_RDWR | O_NOCTTY | O_NONBLOCK, 0); memset(&options, 0x00, sizeof(options)); options.c_cflag |= (CLOCAL | CREAD); options.c_cflag &= ~PARENB; options.c_cflag &= ~CSTOPB; options.c_lflag &= ~(ICANON | ECHO | ECHOE | ISIG); options.c_oflag &= ~OPOST; options.c_iflag &= ~(IXON | IXOFF | IXANY); options.c_cc[VMIN] = 0; options.c_cc[VTIME] = 0; Javier I. Acosta M. 2013-02-20T22:21:34Z http://community.qnx.com/sf/go/post99292 We have a system running QNX 6.5.0 as a TCP/IP server processing proprietary data. The service threads run with round-robin scheduling. Under very heavy loading, we can get one thread blocking in a recv call on its socket. When this happens, the other service threads run, but the blocked thread prevents any lower-priority threads from running. When this happens, we stop all requests and let the machine catch up. The blocked thread remains in this state. I would have thought that a blocked recv call would suspend, letting lower-priority threads still run. I'm still gathering information on the state of the machine and can post more details later, but I would like to know if it is normal or know condition that a blocked recv call can prevent lower-priority threads from running? Robert Murrell 2013-02-14T15:29:28Z http://community.qnx.com/sf/go/post99272 Hi ! Driver now works good. Maybe I didn't copy it correctly. The media is autodetected and I can communicate with qconn perfectly. Thank you very much ! Iván Cayón 2013-02-14T07:25:11Z http://community.qnx.com/sf/go/post99249 What is is sloginfo? On 2013-02-13 6:42 AM, "Iván Cayón" <community-noreply@qnx.com> wrote: >Hi !! > >Sorry but I get an error loading the library > >Unable to load devn-e1000.so > >I the cksum is correct, I give execute permissions also.. any idea please? > >I do: > >io-pkt-v4 -de1000 -ptcpip > >Thanks in advance > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post99248 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-02-13T12:23:58Z http://community.qnx.com/sf/go/post99248 Hi !! Sorry but I get an error loading the library Unable to load devn-e1000.so I the cksum is correct, I give execute permissions also.. any idea please? I do: io-pkt-v4 -de1000 -ptcpip Thanks in advance Iván Cayón 2013-02-13T11:42:58Z http://community.qnx.com/sf/go/post99189 <font face="Default Sans Serif,Verdana,Arial,Helvetica,sans-serif" size="2">I have an Actiontec Q2000 DSL modem, with a QNX system connected to the internal network. I set up the modem to forward ports 22 & 80 to the IP address of the QNX system. Neither ssh or web is visible externally when set up this way. I also have a Linux box on the same network, and if I forward the same ports to that machine, I can do ssh and web externally. On the internal network, I can ssh and do web to the QNX system. I allow port forwarding in the sshd_config file (which should not matter). I am running QNX 6.4.1 on a custom ARM system. Any help would be appreciated.<div><br></div><div>Thanks,</div><div>Todd Peterson</div><div></div></font> Todd Peterson 2013-02-08T21:50:56Z http://community.qnx.com/sf/go/post99175 You're probably looking for getifaddrs(). Gervais Mulongoy 2013-02-08T14:54:20Z http://community.qnx.com/sf/go/post99157 Hi ! Thank you so much !! :) Iván Cayón 2013-02-08T08:09:58Z http://community.qnx.com/sf/go/post99155 Hello, The network adapter can be registered as "eth0", "rt0", etc. ... How can I determine a device name which the use of C code ? Thank you, Jacek Jacek Rudnicki 2013-02-08T06:45:49Z http://community.qnx.com/sf/go/post99105 You can find an updated driver here http://community.qnx.com/sf/wiki/do/viewPage/projects.bsp/wiki/Experimental DriversAndUtilities On 2013-02-06 2:48 AM, "Iván Cayón" <community-noreply@qnx.com> wrote: >Hi ! > > I'm using Intel 82574L Ethernet card loading "e1000" driver, the card is >detected correctly, but I always get "media: Ethernet none", so although >I configure network parameters I can't communicate. I search the problem >and I do: > >ifconfig wm0 media 10BaseT > >And after that, I get "media: Ethernet 10BaseT (full-duplex)", and >network works ok. > >So I thought to put this line in my "/etc/secondary.sh" script to do >every time the board boots. But It's not working >running in the init script, only works when I write the command in the >shell. > >Can you help me please? > >Thanks in advance, >Regards > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post99100 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-02-06T12:33:32Z http://community.qnx.com/sf/go/post99100 Hi ! I'm using Intel 82574L Ethernet card loading "e1000" driver, the card is detected correctly, but I always get "media: Ethernet none", so although I configure network parameters I can't communicate. I search the problem and I do: ifconfig wm0 media 10BaseT And after that, I get "media: Ethernet 10BaseT (full-duplex)", and network works ok. So I thought to put this line in my "/etc/secondary.sh" script to do every time the board boots. But It's not working running in the init script, only works when I write the command in the shell. Can you help me please? Thanks in advance, Regards Iván Cayón 2013-02-06T07:48:51Z http://community.qnx.com/sf/go/post99021 The driver was padding packets to 64 bytes instead of 60 bytes. Glad it is working for you. On 2013-02-02 4:59 AM, "Fei Su" <community-noreply@qnx.com> wrote: >My problem is solved! thank for ur help. Would u mind telling me what's >wrong in the driver? > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post99020 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-02-02T19:52:21Z http://community.qnx.com/sf/go/post99020 My problem is solved! thank for ur help. Would u mind telling me what's wrong in the driver? Fei Su 2013-02-02T09:59:50Z http://community.qnx.com/sf/go/post98999 Thx ,i will try it next monday . Fei Su 2013-02-01T17:11:44Z http://community.qnx.com/sf/go/post98991 Hello. I am implementing UDP communication with other system, QNX is to be server in the communication waiting for requests from clients. The "server" is supposed to be in "standby" mode, so the active one of the two servers will have assigned IP alias. Clients will send UDP packets to the alias IP. So far, so good. The server is able to receive packet via recvmsg(), via msghdr and IP_RECVDSTADDR I am able to find which interface received the packet. But the hard part is to send the packet from given IP address. According to information in Internet one should use either IP_SENDSRCADDR or IP_PKTINFO in msghdr in sendmsg, but none of these is defined in QNX. Is there any way to accomplish what I need to? Thank you. Pavol Kycina 2013-02-01T13:51:56Z http://community.qnx.com/sf/go/post98990 I think that I found the problem. Please try the attached driver. On 2013-02-01 3:30 AM, "Fei Su" <community-noreply@qnx.com> wrote: >Hi the conclusion above is wrong. Just now, I use the fpga to analyse >the frame ,found that qnx pad the frame to 68 bytes (include the fcs). >The fsc error is mentioned by wireshark just because the impc protocl. So >please help me to check why qnx padding more 4 bytes? > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post98983 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-02-01T13:27:25Z http://community.qnx.com/sf/go/post98989 What ping command were you using when you captured this? You also didn't post the output from 'pci -v'. On 2013-02-01 1:16 AM, "Fei Su" <community-noreply@qnx.com> wrote: >hello I have tried the command in qnx and wince ,and the result is not >the same. > >this is capture by the wireshark on qnx: > > > > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post98980 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-02-01T13:05:52Z http://community.qnx.com/sf/go/post98983 Hi the conclusion above is wrong. Just now, I use the fpga to analyse the frame ,found that qnx pad the frame to 68 bytes (include the fcs). The fsc error is mentioned by wireshark just because the impc protocl. So please help me to check why qnx padding more 4 bytes? Fei Su 2013-02-01T08:30:58Z http://community.qnx.com/sf/go/post98982 Because the frame from qnx is fcs error ,and the wireshark put the fsc to me ,so we can see 4 bytes more than the frame from wince. But It seems like the 4byte are not fcs but padding.I guess that if we send frame less than 64byte(include fcs) the driver will send the 4 padding data instead of the fsc. Fei Su 2013-02-01T06:25:18Z http://community.qnx.com/sf/go/post98981 hello I have tried the command in qnx and wince ,and the result is not the same. this is capture by the wireshark on wince: Fei Su 2013-02-01T06:17:28Z http://community.qnx.com/sf/go/post98980 hello I have tried the command in qnx and wince ,and the result is not the same. this is capture by the wireshark on qnx: Fei Su 2013-02-01T06:16:48Z http://community.qnx.com/sf/go/post98947 Ok thx I will try it tomorrow. Fei Su 2013-01-31T15:19:26Z http://community.qnx.com/sf/go/post98946 Actually, please try a "ping -s10 ip.address" and see if that works. On 2013-01-31 10:15 AM, "Fei Su" <community-noreply@qnx.com> wrote: >I am pretty sure that it will padding ,but I'm off duty,so I can try the >ping command tomorrow. > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post98944 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-01-31T15:16:57Z http://community.qnx.com/sf/go/post98945 I am pretty sure that it will padding ,but I'm off duty,so I can try the ping command tomorrow. Fei Su 2013-01-31T15:16:17Z http://community.qnx.com/sf/go/post98944 I am pretty sure that it will padding ,but I'm off duty,so I can try the ping command tomorrow. Fei Su 2013-01-31T15:15:15Z http://community.qnx.com/sf/go/post98942 I have checked the rtl driver and it is padding the packets, as I can ping with 10-byte packets. There might be a different version of the chip that doesn't pad the packets. On 2013-01-31 9:42 AM, "Dave Brown (QNX)" <community-noreply@qnx.com> wrote: >This is the Ethernet minimum . It could be that one driver is padding out >the packet, while the other is not. Some of the drivers maintain specific >statistics for short packets being dropped. > >Dave > >-----Original Message----- >From: Fei Su [mailto:community-noreply@qnx.com] >Sent: January-30-13 9:58 PM >To: general-networking >Subject: EtherCat Problem > >Hi , > I'm using bpf to send and receive the EtherCat frame,but there are >something strange than when I send a broadcast ECAT frame and this frame >is less then 60 bytes,I can't receive the frame form the slave, but I add >a command 'nop' to fill the frame size over 60 bytes ,I can receive from >slave. I am using QNX6.5 and the ethernet card rtl8139. And also when I >use rtl8169,there is no problem like this. > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post98914 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com > > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post98935 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2013-01-31T15:07:14Z http://community.qnx.com/sf/go/post98935 This is the Ethernet minimum . It could be that one driver is padding out the packet, while the other is not. Some of the drivers maintain specific statistics for short packets being dropped. Dave -----Original Message----- From: Fei Su [mailto:community-noreply@qnx.com] Sent: January-30-13 9:58 PM To: general-networking Subject: EtherCat Problem Hi , I'm using bpf to send and receive the EtherCat frame,but there are something strange than when I send a broadcast ECAT frame and this frame is less then 60 bytes,I can't receive the frame form the slave, but I add a command 'nop' to fill the frame size over 60 bytes ,I can receive from slave. I am using QNX6.5 and the ethernet card rtl8139. And also when I use rtl8169,there is no problem like this. _______________________________________________ General http://community.qnx.com/sf/go/post98914 To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Dave Brown 2013-01-31T14:42:21Z http://community.qnx.com/sf/go/post98920 The driver or mac will padding for me or even I am padding it myself ,the problem is still occur. Fei Su 2013-01-31T08:49:13Z http://community.qnx.com/sf/go/post98919 Hi, EtherCAT is based on STANDARD Ethernet media ... so you shouldn't use packets smaller than the minimum packet size. --Armin http://www.steinhoff-automation.com Fei Su wrote: > Hi , > I'm using bpf to send and receive the EtherCat frame,but there are something strange than when I send a broadcast ECAT frame and this frame is less then 60 bytes,I can't receive the frame form the slave, but I add a command 'nop' to fill the frame size over 60 bytes ,I can receive from slave. I am using QNX6.5 and the ethernet card rtl8139. And also when I use rtl8169,there is no problem like this. > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post98914 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com > Armin Steinhoff 2013-01-31T08:34:01Z http://community.qnx.com/sf/go/post98914 Hi , I'm using bpf to send and receive the EtherCat frame,but there are something strange than when I send a broadcast ECAT frame and this frame is less then 60 bytes,I can't receive the frame form the slave, but I add a command 'nop' to fill the frame size over 60 bytes ,I can receive from slave. I am using QNX6.5 and the ethernet card rtl8139. And also when I use rtl8169,there is no problem like this. Fei Su 2013-01-31T02:58:14Z http://community.qnx.com/sf/go/post98881 Thanks Sean! I'll give it a try in the morning. Oleg V 2013-01-29T21:25:58Z http://community.qnx.com/sf/go/post98880 On Tue, Jan 29, 2013 at 04:06:37PM -0500, Oleg V wrote: > Thanks Sean > > Interesting, I just checked again io-pkt docs at http://www.qnx.com/developers/docs/6.4.0/neutrino/utilities/i/io-pkt.html and seems can't find any mention you can load more than one driver at a time. In first message you mentioned 6.5 sp1. Docs are here: http://www.qnx.com/developers/docs/6.5.0_sp1/index.jsp?topic=%2Fcom.qnx.doc.neutrino_utilities%2Fi%2Fio-pkt.html > > Just to confirm, are you saying I can simply do: > > io-pkt-v4-hc -vvvv -d<driver1> -d<driver2> -ptcpip > ifconfig en0 ... > ifconfig en1 ... > > Is that correct? Right. Sean Boudreau 2013-01-29T21:10:04Z http://community.qnx.com/sf/go/post98879 Thanks Sean Interesting, I just checked again io-pkt docs at http://www.qnx.com/developers/docs/6.4.0/neutrino/utilities/i/io-pkt.html and seems can't find any mention you can load more than one driver at a time. Just to confirm, are you saying I can simply do: io-pkt-v4-hc -vvvv -d<driver1> -d<driver2> -ptcpip ifconfig en0 ... ifconfig en1 ... Is that correct? Thanks Oleg V 2013-01-29T21:06:37Z http://community.qnx.com/sf/go/post98878 Sounds like you simply want: # io-pkt -d driver1 -d driver2 If you want multiple stacks you'll have do add the 'prefix' option to target them individually with utilities. # io-pkt -d driver1 # ifconfig # io-pkt -d driver2 -ptcpip prefix=/alt # SOCK=/alt ifconfig This is all be documented in the io-pkt docs. On Tue, Jan 29, 2013 at 03:53:48PM -0500, Oleg V wrote: > Greetings QNX community > > Our QNX 6.5.0 SP1 project requires two Ethernet connections, both using USB Ethernet style adapters. Although the adapters are of the same nature, they are different mfg/part numbers and hence use different QNX drivers. > > Both devices are successfully enumerated by the HC (as listed by usb tool) and I can get each connection to work separately using io-pkt-v4-hc/ifconfig harness: > > io-pkt-v4-hc -vvvv -d<driver> -ptcpip > ifconfig en0 <ip> netmask <mask> > > The problem starts when I'm trying to get both interfaces up at the same time: > > io-pkt-v4-hc -vvvv -d<driver1> -ptcpip <-- Success, /dev/io-net/en0 appears > io-pkt-v4-hc -vvvv -d<driver2> -ptcpip <-- Success, but en1 doesn't appear anywhere as i was expecting. No errors reported though. > > Then I read carefully again through io-pkt-v4-hc documentation and found something called "instance" in there. So I tried specifying different instance numbers for each adapter: > > io-pkt-v4-hc -i 1 -vvvv -d<driver1> -ptcpip <-- Success, /dev/io-net1/en0 appears > io-pkt-v4-hc -i 2 -vvvv -d<driver2> -ptcpip <-- Success, /dev/io-net2/en0 appears > > At this point I'm very confused, this is clearly is not what I want to achieve. This seems to be creating separate stacks and further I don't quite see how to use ifconfig in this setup as both interfaces are registered as "en0". > > I also tried to play with "enmap" parameter but this one seems to be ignored completely. > > All I want is just to get my two network interfaces under /dev/io-net/ as > > /dev/io-net/en0 > /dev/io-net/en1 > > So that I can configure them as usual with > > ifconfig en0 <ip1> netmask <mask1> > ifconfig en1 <ip2> netmask <mask2> > > > Any suggestions? > > Thanks. > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post98877 > To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com Sean Boudreau 2013-01-29T21:00:06Z http://community.qnx.com/sf/go/post98877 Greetings QNX community Our QNX 6.5.0 SP1 project requires two Ethernet connections, both using USB Ethernet style adapters. Although the adapters are of the same nature, they are different mfg/part numbers and hence use different QNX drivers. Both devices are successfully enumerated by the HC (as listed by usb tool) and I can get each connection to work separately using io-pkt-v4-hc/ifconfig harness: io-pkt-v4-hc -vvvv -d<driver> -ptcpip ifconfig en0 <ip> netmask <mask> The problem starts when I'm trying to get both interfaces up at the same time: io-pkt-v4-hc -vvvv -d<driver1> -ptcpip <-- Success, /dev/io-net/en0 appears io-pkt-v4-hc -vvvv -d<driver2> -ptcpip <-- Success, but en1 doesn't appear anywhere as i was expecting. No errors reported though. Then I read carefully again through io-pkt-v4-hc documentation and found something called "instance" in there. So I tried specifying different instance numbers for each adapter: io-pkt-v4-hc -i 1 -vvvv -d<driver1> -ptcpip <-- Success, /dev/io-net1/en0 appears io-pkt-v4-hc -i 2 -vvvv -d<driver2> -ptcpip <-- Success, /dev/io-net2/en0 appears At this point I'm very confused, this is clearly is not what I want to achieve. This seems to be creating separate stacks and further I don't quite see how to use ifconfig in this setup as both interfaces are registered as "en0". I also tried to play with "enmap" parameter but this one seems to be ignored completely. All I want is just to get my two network interfaces under /dev/io-net/ as /dev/io-net/en0 /dev/io-net/en1 So that I can configure them as usual with ifconfig en0 <ip1> netmask <mask1> ifconfig en1 <ip2> netmask <mask2> Any suggestions? Thanks. Oleg V 2013-01-29T20:53:48Z http://community.qnx.com/sf/go/post98211 Hello, I'm new to QNX,and I look into the io-pkt architecture .Are there some api for me to use the "ether iput"? I just want to transfer the data via the ethernet without the protocol such as tcpip. fei su 2012-12-18T16:02:00Z http://community.qnx.com/sf/go/post97945 Thanks for the reply. Already tried this. Thought we were communicating properly, but were suspicious. Took the wire out and discovered they were going through the stack. We need to send packets over a wire. Simple connectivity is not enough. Michael Wicks 2012-12-10T16:31:54Z http://community.qnx.com/sf/go/post97930 It is possible but there are conditions, for example if you have two interfaces each configured with an IP address that happen to be on the same subnet, while you will be able to receive communication on both interfaces transmission will work from only one interface since as you can't have duplicate entries in the routing table. In such a case, you would need an instance of io-pkt per interface so you have two routing tables. You could get away with creating an alias, so giving an interface a second IP address and this can be on the same subnet. When you transmit the packets aren't going to get sent over the wire (so layer 2) they get forwarded from one queue to the next within the stack. Depending on your requirements though this might not be sufficient for a self test. Wouldn't pinging another host on the network be a sufficient test to demonstrate connectivity? Gervais Mulongoy 2012-12-10T13:57:43Z http://community.qnx.com/sf/go/post97929 Is it possible for two IP addresses to communicate with each other in the same network on the same machine? Michael Wicks 2012-12-10T13:54:09Z http://community.qnx.com/sf/go/post97803 Hi, we just loose hours searching a problem in our ethercat filter (lsm-ecat.so) with following result: 6.5.0 io-pkt is complied without "ALTQ". 6.5.0SP1 io-pkt is complied with "ALTQ". Why is ALTQ not defined per default in SP1? If so, recompiling an lsm under 6.5.0 SPI would show up the changes at once. Following release note "The io-pkt* manager now supports the ALTQ feature. You can use pfctl to manage it. (Ref# 107832, 142280) " does not tell you: "Recompile your LSM's with -DALTQ. We have changed the binary Interface!" Regards Michael Michael Tasche 2012-12-05T10:51:37Z http://community.qnx.com/sf/go/post97745 You will need to use formal support channels in order to have this resolved. From: mario sangalli <community-noreply@qnx.com<mailto:community-noreply@qnx.com>> Reply-To: <general-networking@community.qnx.com<mailto:general-networking@community.qnx.com>> Date: Tue, 4 Dec 2012 02:41:55 -0500 To: general-networking <general-networking@community.qnx.com<mailto:general-networking@community.qnx.com>> Subject: Slinger trouble on 6.5 Hi, I've a few web pages using slinger. All works fine on 6.4, but the same application crashes under QNX6.5 : I've tested the same apps and slinger under a x86 system and under a ppc system, with same error: sloginfo -c Time Sev Major Minor Args Nov 22 18:22:46 5 21 0 run fault pid 466961 tid 2 signal 11 code 1 ip 0xfe3657c4 usr/sbin/slinger Here the slinger start up script: #! /bin/ksh export HTTPD_ROOT_DIR=/usr/web/doc export HTTPD_ROOT_DOC=test.shtml export HTTP_SCRIPALIAS=/usr/web/cgi slinger -d -s -e & Maybe the binary included in the QNX6.5 are not updated or recompiled to match OS version? Tx M. Sangalli _______________________________________________ General http://community.qnx.com/sf/go/post97744 To cancel your subscription to this discussion, please e-mail general-networking-unsubscribe@community.qnx.com<mailto:general-networking-unsubscribe@community.qnx.com> Mate Szarvas 2012-12-04T08:12:47Z http://community.qnx.com/sf/go/post97744 Hi, I've a few web pages using slinger. All works fine on 6.4, but the same application crashes under QNX6.5 : I've tested the same apps and slinger under a x86 system and under a ppc system, with same error: sloginfo -c Time Sev Major Minor Args Nov 22 18:22:46 5 21 0 run fault pid 466961 tid 2 signal 11 code 1 ip 0xfe3657c4 usr/sbin/slinger Here the slinger start up script: #! /bin/ksh export HTTPD_ROOT_DIR=/usr/web/doc export HTTPD_ROOT_DOC=test.shtml export HTTP_SCRIPALIAS=/usr/web/cgi slinger -d -s -e & Maybe the binary included in the QNX6.5 are not updated or recompiled to match OS version? Tx M. Sangalli mario sangalli 2012-12-04T07:41:55Z http://community.qnx.com/sf/go/post96886 We don't have any optical Broadcom adapters here, so if you would like to send your hardware, we can take a look at it. On 2012-11-06 8:27 AM, "a b" <community-noreply@qnx.com> wrote: >Hi there, > >We have a problem with Allied Telesis AT-2916SX optical network adapter >with BCM5703X chipset on it. My QNX 6.5 Neutrino automatically loads >drivers at startup. Almost everything seems to be ok except media field >displayed after typing ifconfig: >media: Ethernet 100BaseTX. >I think it should be 100 or 1000BaseSX. >We checked our cards under linux and they work correctly. >Of course under QNX they don't. > >Can anyone help us with our problem? > > > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post96885 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2012-11-06T13:30:28Z http://community.qnx.com/sf/go/post96885 Hi there, We have a problem with Allied Telesis AT-2916SX optical network adapter with BCM5703X chipset on it. My QNX 6.5 Neutrino automatically loads drivers at startup. Almost everything seems to be ok except media field displayed after typing ifconfig: media: Ethernet 100BaseTX. I think it should be 100 or 1000BaseSX. We checked our cards under linux and they work correctly. Of course under QNX they don't. Can anyone help us with our problem? a b 2012-11-06T13:27:21Z http://community.qnx.com/sf/go/post95872 We also have a similar problem on or ppc freescale P1022 (dual core) With one interface, no porblem as long as we enable the 2nd interface , when we stress our server program, after minuts the first io-pkt-v4-hc starts consumming 100% CPU of core 1 after a while the io-pkt-v4-hc of the 2nd interface consummes 100% of the core 2 And it's not a problem of conflicting addresses of the interfaces (subnet are totally distinct...) Pierre BODIN 2012-09-27T11:25:42Z http://community.qnx.com/sf/go/post95247 Dear Hugh Thanks for your response. Because problem occurs rarely, I put the board under test, as soon as it fails, I'll report the sloginfo messages. following is ifconfig output. # ifconfig lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 33192 inet 127.0.0.1 netmask 0xff000000 en0: flags=80008943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST,SHIM> mtu 1500 address: 00:65:43:21:22:60 media: Ethernet 100baseTX full-duplex status: active inet 192.168.100.226 netmask 0xffffff00 broadcast 192.168.100.255 en1: flags=80008943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST,SHIM> mtu 1500 address: 00:65:43:21:22:61 media: Ethernet none full-duplex status: active inet 192.168.101.226 netmask 0xffffff00 broadcast 192.168.101.255 # Regards antony Lewis 2012-08-29T08:08:00Z http://community.qnx.com/sf/go/post95190 What is the output of nicinfo when this happens? Are there any error messages in sloginfo? We will also need the output from "ifconfig". On 2012-08-28 6:01 AM, "antony Lewis" <community-noreply@qnx.com> wrote: >Dear all > >We have QNX 6.4.1 running on a customized board based on PPC440ep >processor, using customized version of AMCC PPC440 EP/GR BSP. Board has >two ethernet interfaces on different subnet . > >Rarely we have problem with ethernet. Board not detected by pinging it >through PC (ping fails), but PC can detected through board using ping >command. > >Both PC and board are connected to one switch. > >Is there any idea about this problem? > >Thanks > > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post95187 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2012-08-28T11:44:10Z http://community.qnx.com/sf/go/post95187 Dear all We have QNX 6.4.1 running on a customized board based on PPC440ep processor, using customized version of AMCC PPC440 EP/GR BSP. Board has two ethernet interfaces on different subnet . Rarely we have problem with ethernet. Board not detected by pinging it through PC (ping fails), but PC can detected through board using ping command. Both PC and board are connected to one switch. Is there any idea about this problem? Thanks antony Lewis 2012-08-28T10:01:57Z http://community.qnx.com/sf/go/post94837 So, we have a pretty old version of QNX (6.2.0) which we have been using for quite some time. We have a multi-platform C++ OS interface for all of our common OS calls, socket io, file io, ipc, timers, the works. We have been using this for many years, so many of the bugs have been worked out pretty well. So, we have a bug which we have only seen in the QNX side of things which we have tried to address on a couple of occasions. We have two programs running on the same board, one which drives the system, and one which simulates some IO using a socket interface which injects signal values into our drive software. The rate which we push code into the socket is pretty high (25 Hz or so), and the messages are what I would consider medium-sized (maybe 2K?). We have not seen this issue when running our drive software, just in simulation. What seems to happen, is the drive software stops servicing the socket to the simulation. The simulation socket send queue fills up until we run out of memory, and causes the application to lock up the board. We have implemented several "fixes". I am able to detect the filling of the send queue, so I have tried closing the socket. This generates a broken pipe on the simulation side, and sometimes the system is able to recover, and sometimes it locks up the board. It usually takes several hours for the anomaly to occur. This issue could be causes by a number of things (a mutex/semaphore deadlock is suspect on the drive software perhaps, or maybe we are running out of other resources, like socket file handles?). I can add some debug on the simulation side, but I am looking for a good way to debug the issue by adding calls when I detect the send queue filling on the simulation side. What things should I look at? Is there any way to detect a mutex/semaphore deadlock from outside the application? I obviously can't add much debug in the semaphore/mutex code without generating a ton of debug, since it takes hours to reproduce. So, what I am looking for, would be good things to check from the simulation software, or otherwise. Once we detect the condition, we have more than a few seconds (maybe a minute) with which we can debug before the board becomes non-functional. Thanks Kevin White 2012-08-15T12:37:50Z http://community.qnx.com/sf/go/post94326 Hi, We are looking to create a forum where RIM (i.e. Wi-Fi) and Vendor (i.e. TI) resources can have discussions. Is there a person I can give a quick call and have a chat with, on this topic? Reason being, I am not sure how many people will be receiving this email. Thanks, Rizwan Malik Project Management, Networking Technologies Research In Motion Limited Office: (519) 888-7465 x10558 Mobile: (519) 573-2802 [Description: C:\Documents and Settings\mumalik\Application Data\Microsoft\Signatures\bblogo.gif] Rizwan Malik 2012-07-18T17:58:39Z http://community.qnx.com/sf/go/post93724 You cannot have 2 interfaces on the same subnet. If you want to do this, you will have to start 2 instances of io-pkt. -- Hugh Brown QNX Software Systems Limited 1001 Farrar Rd., Ottawa. ON. K2K 0B3. Telephone: 613-591-0931 On 12-06-15 3:56 AM, "Marcin Kajzler" <community-noreply@qnx.com> wrote: >Hi, >I am using Bernecker computer with two integrated ethernet interfaces. >When I connect this B computer to the switch like in attachment. Ping >both of interfaces from PC computer, there is a strange behaviour of the >system. When I remove LNK_2 interface en1 stops to respond for ping. When >I remove LNK_1 connected to en0 with connected LNK_2 both interfaces stop >to respond for pings. After restore of LNK_1 both interfaces star to >respond for pings. >I do not have any clue what is going on. With both connected links >connections work stable. >Regards, Marcin > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post93688 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2012-06-18T12:24:26Z http://community.qnx.com/sf/go/post93688 Hi, I am using Bernecker computer with two integrated ethernet interfaces. When I connect this B computer to the switch like in attachment. Ping both of interfaces from PC computer, there is a strange behaviour of the system. When I remove LNK_2 interface en1 stops to respond for ping. When I remove LNK_1 connected to en0 with connected LNK_2 both interfaces stop to respond for pings. After restore of LNK_1 both interfaces star to respond for pings. I do not have any clue what is going on. With both connected links connections work stable. Regards, Marcin Marcin Kajzler 2012-06-15T07:56:16Z http://community.qnx.com/sf/go/post92953 If you have the BSP the source code for the network driver should be included with it. If you have QNX Momentics development suite, the network documentation should be included. -- Hugh Brown QNX Software Systems Limited 1001 Farrar Rd., Ottawa. ON. K2K 0B3. Telephone: 613-591-0931 On 12-05-10 2:02 PM, "shivarama bhat" <community-noreply@qnx.com> wrote: >Hi, >thank you very much. >Is Network driver part of BSP or part of io_pkt? >Where can I find documentation on the APIs of io_pkt used for developing >BSP. >Currently we dont have acess to network project or source code of core >networking. >Is it possible to change PHY driver without access to source code of core >networking? >regards >Shivaram > > > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post92950 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2012-05-11T06:41:32Z http://community.qnx.com/sf/go/post92950 Hi, thank you very much. Is Network driver part of BSP or part of io_pkt? Where can I find documentation on the APIs of io_pkt used for developing BSP. Currently we dont have acess to network project or source code of core networking. Is it possible to change PHY driver without access to source code of core networking? regards Shivaram shivarama bhat 2012-05-10T18:02:40Z http://community.qnx.com/sf/go/post92932 The PHY drivers as part of the network driver, so you will have to modify the network driver for different PHY modes. -- Hugh Brown QNX Software Systems Limited 1001 Farrar Rd., Ottawa. ON. K2K 0B3. Telephone: 613-591-0931 On 12-05-09 1:30 AM, "shivarama bhat" <community-noreply@qnx.com> wrote: >We are using Freescale P1020rdb_PA version of the board and BSP. However >we want to use different PHYs and mode of each ethernet port available on >P1020. Where can I find the documentation and source code for introducing >new PHY drivers, configuring the ether port modes such as RMII, SGMII etc. > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post92931 >To cancel your subscription to this discussion, please e-mail >general-networking-unsubscribe@community.qnx.com Hugh Brown 2012-05-09T09:23:08Z http://community.qnx.com/sf/go/post92931 We are using Freescale P1020rdb_PA version of the board and BSP. However we want to use different PHYs and mode of each ethernet port available on P1020. Where can I find the documentation and source code for introducing new PHY drivers, configuring the ether port modes such as RMII, SGMII etc. shivarama bhat 2012-05-09T05:30:23Z http://community.qnx.com/sf/go/post92699 I'm trying to send messages between a few computers and would like to use name_open but the name doesn't appear to propagate across nodes with different endianess. I have 2 little endian nodes (x86, armle-v7) and one big endian node (ppcbe) all running QNX 6.5.0. They all have qnet and gns running with the x86 node acting as the gns server. Although the nodes can see each other on the network ( i.e. 'ls /net' from any node shows all 3 ) in /dev/name/global/net they can only ever see the like-endian nodes. Is this a limitation of gns? Is there a way around this? Thanks. Steven Lougheed 2012-04-23T15:15:13Z http://community.qnx.com/sf/go/post92626 I have tried export http_proxy=..... but have had no luck with that yet. Richard Elswick 2012-04-18T21:21:42Z http://community.qnx.com/sf/go/post92625 How do I setup a network proxy within QNX 6.5 to access the internet? Richard Elswick 2012-04-18T21:20:38Z http://community.qnx.com/sf/go/post92469 Hi, An Alstom customer is asking if there is a port of SNTP for 6.5.0, or if our ntpd supports SNTP? (I cannot find any mention of SNTP in our docs, but I can find references to SNTP in SVN under io-pkt/dist.) Regards, Garry Garry Bleasdale 2012-04-05T11:49:28Z http://community.qnx.com/sf/go/post92213 I am currently on a SNMP project. My question is how to compile my SNMPv1 MIB files and generate C-language code for my SNMP agent. I am using QNX 6.5 SDP. It looks QNX SMIC only can generate MOSY format, but can not create C-language skeleton code for the agent. Does QNX6.5 have a tool for that? I know QNX had an Extended Networking TDK for QNX6.3, can I use that one for QNX6.5 to generate C code from MIB file? Thanks a lot! Yue Li 2012-03-20T18:21:11Z http://community.qnx.com/sf/go/post92058 Here's the latest on this: http://community.qnx.com/sf/wiki/do/viewPage/projects.community/wiki/UpdatedQNXSourceAccessPolicyFAQ Regards, -seanb On Mon, Mar 12, 2012 at 03:15:20PM -0400, Greg Ludewig wrote: > I'm trying to download the networking source. Here's what I get when I attempt to follow the instructions: > > svn checkout --username gludewig@minnetronix.com http://community.qnx.com/svn/repos/core_networking/trunk > Authentication realm: <http://community.qnx.com:80> Authorization Realm > Password for 'gludewig@minnetronix.com' : > svn: access to 'svn/repos/core_networking/trunk' forbidden > > I am a member of the Networking project, and I verified that I'm entering my userid and password correctly. > > Can someone help me so I can download the networking source code? > > Thanks! > > Greg Ludewig > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post92057 > Sean Boudreau 2012-03-12T19:21:19Z http://community.qnx.com/sf/go/post92057 I'm trying to download the networking source. Here's what I get when I attempt to follow the instructions: svn checkout --username gludewig@minnetronix.com http://community.qnx.com/svn/repos/core_networking/trunk Authentication realm: <http://community.qnx.com:80> Authorization Realm Password for 'gludewig@minnetronix.com' : svn: access to 'svn/repos/core_networking/trunk' forbidden I am a member of the Networking project, and I verified that I'm entering my userid and password correctly. Can someone help me so I can download the networking source code? Thanks! Greg Ludewig Greg Ludewig 2012-03-12T19:15:18Z http://community.qnx.com/sf/go/post91784 How do I setup a proxy server id in the OS, so it can navigate through my corporate firewall? http_proxy=http://proxy.com OR some other means? It also expects a uersname/password as well. Thanks for the help in advance. Richard Elswick 2012-02-24T23:15:54Z http://community.qnx.com/sf/go/post91707 intSocket is just defined as: int intSocket The error display code is just they're because I know they're is a problem and I wanted to display it. I've moved on a little since the initial post, I've now edit /etc/services and added an entry http 80/tcp Having done this the error returned is now 243, Protocol not configured or not supported So I'm thinking that something more is missing from our build. Simon Platten 2012-02-22T12:28:09Z http://community.qnx.com/sf/go/post91706 What is intSocket? You shouldn't look at errno unless the return from socket() is -1. If it's -1 make sure 'io-net -ptcpip ...' is running. -seanb On Wed, Feb 22, 2012 at 06:03:19AM -0500, Simon Platten wrote: > I'm trying to use a server socket in a process, the code compiles without errors or warnings, however when I run it I get: > > ESRCH No such process > > This is after creating the socket: > > errno = EOK; > intSocket = socket( AF_INET, SOCK_STREAM, 0 ); > intErr = errno; > printf( "ERROR = %s\n", strerror(intErr) ); > > According to the hellp http://www.qnx.com/developers/docs/6.3.0SP3/neutrino/lib_ref/s/socket.html > > This isn't an error listed for this routine. What do I need to do to resolve this problem? > > I've checked /bin, libsocket.so, libsocket.so.2, npm-tcpip-v4.so and npm_tcpip.so are all present. > > Thank you > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post91702 > Sean Boudreau 2012-02-22T12:17:11Z http://community.qnx.com/sf/go/post91702 I'm trying to use a server socket in a process, the code compiles without errors or warnings, however when I run it I get: ESRCH No such process This is after creating the socket: errno = EOK; intSocket = socket( AF_INET, SOCK_STREAM, 0 ); intErr = errno; printf( "ERROR = %s\n", strerror(intErr) ); According to the hellp http://www.qnx.com/developers/docs/6.3.0SP3/neutrino/lib_ref/s/socket.html This isn't an error listed for this routine. What do I need to do to resolve this problem? I've checked /bin, libsocket.so, libsocket.so.2, npm-tcpip-v4.so and npm_tcpip.so are all present. Thank you Simon Platten 2012-02-22T11:03:17Z http://community.qnx.com/sf/go/post91675 Hi, I have problem between two nodes sporadically. I can't list /net/col-ptz-adq2/ filesystem from col-ptz-adq1. (ls /net/col-ptz-adq2/ blocked forever) But if can list /net/col-ptz-adq2/ from col-ptz-dev2. I'm using QNX 6.3.0 #qconfig QNX Installations Installation Name: QNX Momentics 6.3.0 Service Pack 3 Version: 6.3.0 SP3 Base Directory: /usr/qnx630/ QNX_HOST: /usr/qnx630/host/qnx6/x86/ QNX_TARGET: /usr/qnx630/target/qnx6/ Additional Packages Package Name: QNX Neutrino Core OS Version: 6.3.2 Base: QNX Momentics 6.3.0 SP1, SP2 or SP3 Install Path: /usr/qnx630 #hostname col-ptz-adq1 #cat /proc/qnetstats kif net_server : 22,22 kif waiting : 17,18 kif net_client : 1,8 kif buffer : 0,8 kif outbound_msgs : 16,17 kif vtid : 1,8 kif server_msgs : 16,17 kif nd_down : 12 kif nd_up : 20 kif nd_changed : 0 kif send_acks : 0 kif client_kercalls :2587617969 kif server_msgs :1696506283 kif server_unblock : 30 qos tx_begin_errors : 0 qos tx_done_errors : 0 qos tx_throttled : 1 qos tx_failed : 0 qos pkts_rxd_noL4 : 2608 qos tx_conn_created : 17 qos tx_conn_deleted : 12 qos rx_conn_created : 17 qos rx_conn_deleted : 12 qos rx_seq_order : 3 **** Qnet compiled on Jun 20 2006 at 13:43:56 running on col-ptz-adq1 **** Tx Connections: col-ptz-hmi1 i 1 st 3 ln 24 rn 4 lc 17 rc 5 hq 0 tq 0 ns 28 ds 27 col-ptz-db1 i 1 st 3 ln 23 rn 3 lc 14 rc 4 hq 0 tq 0 ns 6 ds 5 col-ptz-dev2 i 1 st 3 ln 21 rn 6 lc 13 rc 7 hq 0 tq 0 ns 249 ds 248 col-ptz-web i 1 st 3 ln 18 rn 6 lc 10 rc 2 hq 0 tq 0 ns 120743 ds 120742 col-ptz-adq2 i 1 st 3 ln 4 rn 29 lc 1 rc 4 hq 0 tq 0 ns 42 ds 41 **** Rx Connections: ln 24 rn 4 lc 17 rc 5 tk 59131708 sq 27 ln 23 rn 3 lc 14 rc 4 tk 59131708 sq 5 ln 21 rn 6 lc 13 rc 7 tk 59131714 sq 242 ln 18 rn 6 lc 10 rc 2 tk 59131692 sq 120689 ln 4 rn 29 lc 1 rc 4 tk 59131723 sq -76081198 **** L4 Status: slot 0 en1 mtu 1496 ack 1 crc 0 c 1 e 1 i 0 txd ok 4235598745 txd bad 0 txd descr 0 txd still 0 tx timeouts 16 tx slow 0 rxd ok 4237626848 rxd bad dr 0 rxd bad L4 0 rxd dropped 0 rxd duplic 0 rxd nacks 0 slot 1 is unused ... ... ... #hostname col-ptz-adq2 #cat /proc/qnetstats kif net_server : 3,10 kif waiting : 2,11 kif net_client : 16,18 kif buffer : 0,6 kif outbound_msgs : 1,10 kif vtid : 16,18 kif server_msgs : 1,10 kif nd_down : 17 kif nd_up : 25 kif nd_changed : 0 kif send_acks : 0 kif client_kercalls :2461467468 kif server_msgs :2038960397 kif server_unblock : 12 qos tx_begin_errors : 0 qos tx_done_errors : 7 qos tx_throttled : 52 qos tx_failed : 11 qos pkts_rxd_noL4 : 2613 qos tx_conn_created : 23 qos tx_conn_deleted : 17 qos rx_conn_created : 23 qos rx_conn_deleted : 17 qos rx_seq_order : 45 **** Qnet compiled on Jun 20 2006 at 13:43:56 running on col-ptz-adq2 **** Tx Connections: col-ptz-hmi1 i 1 st 3 ln 3 rn 5 lc 21 rc 6 hq 0 tq 0 ns 46 ds 45 col-ptz-db2 i 1 st 3 ln 6 rn 5 lc 19 rc 5 hq 0 tq 0 ns 321 ds 320 col-ptz-db1 i 1 st 3 ln 4 rn 2 lc 16 rc 3 hq 0 tq 0 ns 326 ds 325 col-ptz-dev2 i 1 st 3 ln 7 rn 4 lc 15 rc 23 hq 0 tq 0 ns 193 ds 192 col-ptz-web i 1 st 3 ln 1 rn 3 lc 13 rc 8 hq 0 tq 0 ns 173944 ds 173943 col-ptz-adq1 i 1 st 3 ln 29 rn 4 lc 4 rc 1 hq 0 tq 0 ns -76081197 ds -76081198 **** Rx Connections: ln 3 rn 5 lc 21 rc 6 tk 63380465 sq 45 ln 6 rn 5 lc 19 rc 5 tk 63380487 sq 320 ln 4 rn 2 lc 16 rc 3 tk 63380486 sq 325 ln 7 rn 4 lc 15 rc 22 tk 63380472 sq 192 ln 1 rn 3 lc 13 rc 8 tk 63380468 sq 173741 ln 29 rn 4 lc 4 rc 1 tk 63380440 sq -1 **** L4 Status: slot 0 en1 mtu 1496 ack 1 crc 0 c 1 e 1 i 0 txd ok 530431068 txd bad 0 txd descr 0 txd still 0 tx timeouts 203 tx slow 0 rxd ok 532857173 rxd bad dr 0 rxd bad L4 0 rxd dropped 0 rxd duplic 2 rxd nacks 0 slot 1 is unused ... ... ... How can I solve this problem? (without reeboot or umount/mount enX.) Thanks. Gabriel Coteja 2012-02-20T19:14:23Z http://community.qnx.com/sf/go/post91542 Hi, I make a smp boot USB base on QNX6.5 as the bootable USB device. However, the UEFI bios system couldn't install it correctly through USB boot. It always hang up as the attached screen shot (1.jpg). I ever ask the BIOS company (Award) if they support QNX or not? Their answer is : when we saw the string of QNX on the screen, it means our USB device already take care the following process. In addition, we also use a QNX6.5 hard disk to boot the same platform, and it could boot correctly. Any suggestion of how this can be solved? Thanks! HsiangChih Tsai 2012-02-14T00:18:19Z http://community.qnx.com/sf/go/post91506 Sean, thanks for your quick response. I was wondering if we keep more than one thread (on sender and receiver process), for each of the connector-channel pair? any thoughts on having this model? -Hari Haridas Puthiyapurayil 2012-02-13T01:05:14Z http://community.qnx.com/sf/go/post91505 The classic thing to look out for is having both servers SEND blocked on each other (deadlock). You can keep one end the server and have it pass data back in the Reply rather than the Send (eg write() vs read()). Regards, -seanb ----- Original Message ----- From: Haridas Puthiyapurayil [mailto:community-noreply@qnx.com] Sent: Saturday, February 11, 2012 01:55 AM To: general-networking <post91499@community.qnx.com> Subject: Using channel at client and server side for bi-directional communication Hi, I'm trying to use QNX IPC calls (MSgSend/MsgReply) mechnism to have bi-directional communication between two processes. Basically using channel at the both processes. I heard there could be a issue with this design and trying to understand more details on this design? Looks like QNX recommends using send hierarchy and non-blocking pulse. Please help me to identify the issues with the first design as described in the begining of this message? Thanks Hari _______________________________________________ General http://community.qnx.com/sf/go/post91499 Sean Boudreau 2012-02-12T15:28:06Z http://community.qnx.com/sf/go/post91499 Hi, I'm trying to use QNX IPC calls (MSgSend/MsgReply) mechnism to have bi-directional communication between two processes. Basically using channel at the both processes. I heard there could be a issue with this design and trying to understand more details on this design? Looks like QNX recommends using send hierarchy and non-blocking pulse. Please help me to identify the issues with the first design as described in the begining of this message? Thanks Hari Haridas Puthiyapurayil 2012-02-11T06:55:09Z http://community.qnx.com/sf/go/post91240 I work with QNX 6.5 and x86 CPU. inetd is runing . inetd.conf loading successed. output of console is : 49172 reaped status 0x100 someone wants telnet accept,ctrl 3 and output in QNX IDE is host reject connection. gao yuan 2012-01-31T07:32:37Z http://community.qnx.com/sf/go/post90981 Is there some reason that rtsp support cannot be enabled in libcurl? There is no "Curl_handler_rtsp" in the 6.5 version. Dennis Kellly 2012-01-18T17:21:26Z http://community.qnx.com/sf/go/post90975 Dear friends Thank you very much for your helpful reply. Fortunately problem was solved, That was related to a linefeed in /etc/passwd. Regards Antony Lewis antony Lewis 2012-01-18T12:03:35Z http://community.qnx.com/sf/go/post90949 If devc-pty is not running you can get an error like this. Also make sure there is a pseudo device available if it is running as other processes (for example pterm) can consume them. Dave > -----Original Message----- > From: Sean Boudreau [mailto:community-noreply@qnx.com] > Sent: January-16-12 9:10 AM > To: general-networking > Subject: Re: Telnetd Problem > > > telnetd may log more to the syslog. > Set up /etc/syslog.conf and start up syslogd. > > Regards, > > -seanb > > On Sun, Jan 15, 2012 at 03:29:53AM -0500, antony Lewis wrote: > > Dear friends > > > > I customized QNX Neutrino 6.4.0 BSP for the AMCC PPC440 EP/GR (bsp- > nto640-amcc-ppc440ep_gr-evk-trunk) to run on my board that is using > PPC440ep. > > (using QNX Momentics 6.4.1) > > > > There is a problem about running telnetd on my board. when i telnet > my board following messages are shown: > > > > " > > QNX Neutrino (localhost) (ttyp0) > > > > login: root > > No home directory. > > Logging in with home = "/". > > : No such file or directory > > > > Connection closed by foreign host. > > " > > > > Following configuration has been done for telnet > > (according to > http://www.qnx.com/developers/docs/6.4.1/neutrino/utilities/t/telnetd.h > tml): > > > > Following files has been included: > > > > /sbin/io-pkt-v4 > > /lib/dll/devn-ppc405-ep440c.so > > /lib/libsocket.so > > /sbin/devc-pty > > /sbin/route > > /sbin/ifconfig > > /usr/sbin/inetd > > /etc/inetd.conf > > /usr/sbin/telnetd > > /etc/services > > /bin/login > > /etc/termcap > > /usr/lib/terminfo > > /etc/hosts > > /bin/sh Linked to /bin/ksh > > > > Please find other configuration details in attachment. > > > > Any idea about problem will be helpful. > > > > Thanks > > Antony > > > > > > > > > > > > _______________________________________________ > > > > General > > http://community.qnx.com/sf/go/post90938 > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90948 Dave Brown 2012-01-16T15:00:23Z http://community.qnx.com/sf/go/post90948 telnetd may log more to the syslog. Set up /etc/syslog.conf and start up syslogd. Regards, -seanb On Sun, Jan 15, 2012 at 03:29:53AM -0500, antony Lewis wrote: > Dear friends > > I customized QNX Neutrino 6.4.0 BSP for the AMCC PPC440 EP/GR (bsp-nto640-amcc-ppc440ep_gr-evk-trunk) to run on my board that is using PPC440ep. > (using QNX Momentics 6.4.1) > > There is a problem about running telnetd on my board. when i telnet my board following messages are shown: > > " > QNX Neutrino (localhost) (ttyp0) > > login: root > No home directory. > Logging in with home = "/". > : No such file or directory > > Connection closed by foreign host. > " > > Following configuration has been done for telnet > (according to http://www.qnx.com/developers/docs/6.4.1/neutrino/utilities/t/telnetd.html): > > Following files has been included: > > /sbin/io-pkt-v4 > /lib/dll/devn-ppc405-ep440c.so > /lib/libsocket.so > /sbin/devc-pty > /sbin/route > /sbin/ifconfig > /usr/sbin/inetd > /etc/inetd.conf > /usr/sbin/telnetd > /etc/services > /bin/login > /etc/termcap > /usr/lib/terminfo > /etc/hosts > /bin/sh Linked to /bin/ksh > > Please find other configuration details in attachment. > > Any idea about problem will be helpful. > > Thanks > Antony > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90938 Sean Boudreau 2012-01-16T14:09:38Z http://community.qnx.com/sf/go/post90944 Try to include /etc/passwd and /etc/shadow into your build file. Regards, PKY Pavol Kycina 2012-01-16T13:09:02Z http://community.qnx.com/sf/go/post90943 Could you please attach a copy of your build file? Gervais Mulongoy 2012-01-16T12:39:04Z http://community.qnx.com/sf/go/post90938 Dear friends I customized QNX Neutrino 6.4.0 BSP for the AMCC PPC440 EP/GR (bsp-nto640-amcc-ppc440ep_gr-evk-trunk) to run on my board that is using PPC440ep. (using QNX Momentics 6.4.1) There is a problem about running telnetd on my board. when i telnet my board following messages are shown: " QNX Neutrino (localhost) (ttyp0) login: root No home directory. Logging in with home = "/". : No such file or directory Connection closed by foreign host. " Following configuration has been done for telnet (according to http://www.qnx.com/developers/docs/6.4.1/neutrino/utilities/t/telnetd.html): Following files has been included: /sbin/io-pkt-v4 /lib/dll/devn-ppc405-ep440c.so /lib/libsocket.so /sbin/devc-pty /sbin/route /sbin/ifconfig /usr/sbin/inetd /etc/inetd.conf /usr/sbin/telnetd /etc/services /bin/login /etc/termcap /usr/lib/terminfo /etc/hosts /bin/sh Linked to /bin/ksh Please find other configuration details in attachment. Any idea about problem will be helpful. Thanks Antony antony Lewis 2012-01-15T08:29:51Z http://community.qnx.com/sf/go/post90869 Hello, I have a question regarding the socket interface send() method, I have a TCP server process that periodically sends data (several kilobytes ~4 times a second) which seems to me like a small amount of data over local network. This works well but I do see an error after some time (the timing varies a lot, between 10 seconds and 2 days), where the send() call will return -1 with errno set to EAGAIN. According to the documentation (http://www.qnx.com/developers/docs/6.5.0/index.jsp?topic=/com.qnx.doc.neutrino_lib_ref/s/send.html) this should happen when send() is called on a non-blocking socket and the operation would block, but my socket is not set to non-blocking... Any idea what else could cause this error? Given the errno description I would be tempted to retry at least once before giving up, but before doing that I would like to be sure I am not masking a system problem. QNX version is 6.5.0. I also have other processes running (a http server, another tcp connection), so there is a bunch of other processes doing socket communication. Martin Duvanel 2012-01-09T17:09:21Z http://community.qnx.com/sf/go/post90803 Custom driver ioctls should be wrapped in an SIOCSDRVSPEC / SIOCGDRVSPEC ioctl. Regards, -seanb On Thu, Dec 22, 2011 at 12:23:03PM -0500, Yurong Sun wrote: > Hi, > > I want o create wifi driver custom ioctol command. How can I find which group, number I can use without conflicting with the existing ones? > > I searched ioctl.h, ieee80211_ioctl.h, but it is not obvious about the convention of how to define the custom ioctl index. > > thanks, > > Yurong > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90748 > Sean Boudreau 2012-01-03T15:46:54Z http://community.qnx.com/sf/go/post90748 Hi, I want o create wifi driver custom ioctol command. How can I find which group, number I can use without conflicting with the existing ones? I searched ioctl.h, ieee80211_ioctl.h, but it is not obvious about the convention of how to define the custom ioctl index. thanks, Yurong Yurong Sun 2011-12-22T17:23:02Z http://community.qnx.com/sf/go/post90461 Thanks for confirmation Regards Praveen ----- Original Message ----- From: "Mario Charest" <community-noreply@qnx.com> To: "general-networking" <post90460@community.qnx.com> Sent: Wednesday, December 07, 2011 9:38 AM Subject: RE: QNET without TCPIP? Is it possible > > > -----Message d'origine----- > De : Praveen VS [mailto:community-noreply@qnx.com] > Envoyé : 6 décembre 2011 23:02 > À : general-networking > Objet : Re: QNET without TCPIP? Is it possible > > Thanks > So if i donot configure IP , the qnet is not working with IP? > Regards > Praveen > > By default QNET is not working with IP even if TCPIP is running. You have > to specify special option to get qnet to run over IP. > > ----- Original Message ----- > From: "Oleh Derevenko" <community-noreply@qnx.com> > To: "general-networking" <post90446@community.qnx.com> > Sent: Wednesday, December 07, 2011 12:03 AM > Subject: Re: QNET without TCPIP? Is it possible > > >> QNet is started if /etc/system/config/useqnet file is present/readable >> (may be an empty file). >> TCPIP driver is always loaded with io-pkt whether you specify -ptcpip or >> not. It is just up to you if to use it. >> >> >> >> _______________________________________________ >> >> General >> http://community.qnx.com/sf/go/post90446 >> >> > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90459 > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90460 > Praveen VS 2011-12-07T04:10:53Z http://community.qnx.com/sf/go/post90460 -----Message d'origine----- De : Praveen VS [mailto:community-noreply@qnx.com] Envoyé : 6 décembre 2011 23:02 À : general-networking Objet : Re: QNET without TCPIP? Is it possible Thanks So if i donot configure IP , the qnet is not working with IP? Regards Praveen By default QNET is not working with IP even if TCPIP is running. You have to specify special option to get qnet to run over IP. ----- Original Message ----- From: "Oleh Derevenko" <community-noreply@qnx.com> To: "general-networking" <post90446@community.qnx.com> Sent: Wednesday, December 07, 2011 12:03 AM Subject: Re: QNET without TCPIP? Is it possible > QNet is started if /etc/system/config/useqnet file is present/readable > (may be an empty file). > TCPIP driver is always loaded with io-pkt whether you specify -ptcpip or > not. It is just up to you if to use it. > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90446 > > _______________________________________________ General http://community.qnx.com/sf/go/post90459 Mario Charest 2011-12-07T04:08:35Z http://community.qnx.com/sf/go/post90459 Thanks So if i donot configure IP , the qnet is not working with IP? Regards Praveen ----- Original Message ----- From: "Oleh Derevenko" <community-noreply@qnx.com> To: "general-networking" <post90446@community.qnx.com> Sent: Wednesday, December 07, 2011 12:03 AM Subject: Re: QNET without TCPIP? Is it possible > QNet is started if /etc/system/config/useqnet file is present/readable > (may be an empty file). > TCPIP driver is always loaded with io-pkt whether you specify -ptcpip or > not. It is just up to you if to use it. > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post90446 > > Praveen VS 2011-12-07T04:01:29Z http://community.qnx.com/sf/go/post90446 QNet is started if /etc/system/config/useqnet file is present/readable (may be an empty file). TCPIP driver is always loaded with io-pkt whether you specify -ptcpip or not. It is just up to you if to use it. Oleh Derevenko 2011-12-06T18:33:54Z http://community.qnx.com/sf/go/post90435 I just tried running the io-pkt as below slay io-pkt-v4-hc io-pkt-v4-hc -drtl -pqnet & didnot configure IP address. I could get other QNX system listed under /net Hope this is the correct way? Regards Praveen Praveen VS 2011-12-06T10:29:17Z http://community.qnx.com/sf/go/post90432 Hello, Is it possible to have QNET on ethernet without using TCPIP? ( similar to what we had in QNX4 , native networking) Regards Praveen Praveen VS 2011-12-06T07:53:18Z http://community.qnx.com/sf/go/post90297 Dear All, I am new to using QNX and embedded systems. I have TI OMAP3530 BeagleBoard and I want to run the demo for it. But, I don't want to do the same things which have done before Is there anyone who can try to help me how to run a swf file which is located at the link below; http://www.qnx.com/download/feature.html?programid=20707 What I have now, a running QNX Operating system and notthing else. What I want to do : running QNX on the board and then run swf file. In order to reach my aim, what should I do? All the best regards, Seyfettin Seyfettin Sünger 2011-11-25T11:35:18Z http://community.qnx.com/sf/go/post90202 Here is yet another crash like this (gdb) x /2wx 133774728 0x7f93d88: 0x0818b66c 0xeef3f024 (gdb) x /128bx 0x0818b65c 0x818b65c: 0xff 0xff 0xff 0xff 0xff 0xff 0x00 0x15 0x818b664: 0x5d 0x0a 0xdd 0x01 0x82 0x04 0x00 0x00 0x818b66c: 0x2a 0x0b 0x07 0x02 0x00 0x00 0x00 0x00 0x818b674: 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x818b67c: 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x818b684: 0x00 0x00 0xf3 0xee 0x00 0xf0 0xf3 0xee 0x818b68c: 0x00 0x00 0x00 0x00 0x00 0xf0 0xf3 0xee 0x818b694: 0x00 0xf0 0xf3 0xee 0x00 0xf0 0x54 0xff 0x818b69c: 0x00 0xf0 0xbf 0x00 0x00 0xf0 0x67 0x00 0x818b6a4: 0x00 0xf0 0xa5 0xfe 0x00 0xf0 0x87 0xe9 0x818b6ac: 0x00 0xf0 0xf3 0xee 0x00 0xf0 0xf3 0xee 0x818b6b4: 0x00 0xf0 0xf3 0xee 0x00 0xf0 0xf3 0xee 0x818b6bc: 0x00 0xf0 0x57 0xef 0x00 0xf0 0x53 0xff 0x818b6c4: 0x00 0xf0 0x48 0x11 0x00 0xc8 0x4d 0xf8 0x818b6cc: 0x00 0xf0 0x41 0xf8 0x00 0xf0 0x58 0x15 0x818b6d4: 0x00 0xf0 0x39 0xe7 0x00 0xf0 0x59 0xf8 Oleh Derevenko 2011-11-20T10:25:54Z http://community.qnx.com/sf/go/post90198 Hi! This relates to QNX 6.3.0SP3. So, I don't know if in current version things are different or not, but just FYI. I've been investigating a crash of io-net which happened after ~7 days of running an application test. Program terminated on signal 11 with following call stack #0 0xb8216e28 in l4_crc32 () from /usr/qnx630/target/qnx6/x86/lib/dll/npm-qnet-l4_lite.so #1 0xb821cfa4 in l4_rx () from /usr/qnx630/target/qnx6/x86/lib/dll/npm-qnet-l4_lite.so #2 0xb8218435 in ?? () from /usr/qnx630/target/qnx6/x86/lib/dll/npm-qnet-l4_lite.so Looking in l4_crc32 disassembler dump in GDB I could assume it was accessing data via first IOV element. Luckily, the IOV was still available in stack frame (gdb) x /2x 0x07f93d88 0x7f93d88: 0x0807a66c 0xeef3f024 As you can see the base is valid but the size is a junk. So, the question is if QNet extracts some payload sizes from packet data, does the code validate those sizes to not exceed packet boundaries? Here is the dump of first 128 bytes of packet (obtained by IOV element's base) in core file. 0x807a65c: 0x00 0x01 0x05 0x01 0x65 0x28 0x00 0x15 0x807a664: 0x5d 0x0a 0xdd 0x01 0x82 0x04 0x00 0x00 0x807a66c: 0x2a 0x02 0x07 0x00 0x20 0x00 0x16 0x00 0x807a674: 0x33 0x06 0x00 0x00 0x05 0x00 0x00 0x00 0x807a67c: 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x807a684: 0x00 0x00 0xf3 0xee 0x00 0xf0 0xf3 0xee 0x807a68c: 0x00 0x00 0x00 0x00 0x00 0xf0 0xf3 0xee 0x807a694: 0x00 0xf0 0xf3 0xee 0x00 0xf0 0x54 0xff 0x807a69c: 0x00 0xf0 0xbf 0x00 0x00 0xf0 0x67 0x00 0x807a6a4: 0x00 0xf0 0xa5 0xfe 0x00 0xf0 0x87 0xe9 0x807a6ac: 0x00 0xf0 0xf3 0xee 0x00 0xf0 0xf3 0xee 0x807a6b4: 0x00 0xf0 0xf3 0xee 0x00 0xf0 0xf3 0xee 0x807a6bc: 0x00 0xf0 0x57 0xef 0x00 0xf0 0x53 0xff 0x807a6c4: 0x00 0xf0 0x48 0x11 0x00 0xc8 0x4d 0xf8 0x807a6cc: 0x00 0xf0 0x41 0xf8 0x00 0xf0 0x58 0x15 0x807a6d4: 0x00 0xf0 0x39 0xe7 0x00 0xf0 0x59 0xf8 Oleh Derevenko 2011-11-19T12:00:03Z http://community.qnx.com/sf/go/post90184 How can I limit the maximum connections for telnet on QNX 6.4? Say only a client can telnet to a server. Thanks for your reply in advance. jason wang 2011-11-18T07:21:15Z http://community.qnx.com/sf/go/post89403 Sorry, Source code access is no longer available here. You will have to go through your customer service/sales contact. Cheers! /P On 11-10-14 10:26 AM, "Charles Snyder" <community-noreply@qnx.com> wrote: >Could I get a reply here to this request please? > >If I have demonstrated some ignorance in my request - please correct me. > >Thanks, >Charles Snyder > > > >_______________________________________________ > >General >http://community.qnx.com/sf/go/post89398 > Patrik Lahti 2011-10-14T15:03:50Z http://community.qnx.com/sf/go/post89398 Could I get a reply here to this request please? If I have demonstrated some ignorance in my request - please correct me. Thanks, Charles Snyder Charles Snyder 2011-10-14T14:26:37Z http://community.qnx.com/sf/go/post89397 Not currently. The closest is bpf / libpcap. Regards, -seanb On Fri, Oct 14, 2011 at 06:54:42AM -0400, Dennis Kellly wrote: > Does io-pkt support anything comparable to packet_mmap in Linux? Apparently it is a zero-copy capture scheme, useful for snffers. Exists in Linux kernels 2.4, 2.6 and.2.8. > > Description here: > > http://lingrok.org/source/xref/linux-2.6-linus/Documentation/networking/packet_mmap.txt > > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post89393 > Sean Boudreau 2011-10-14T14:04:35Z http://community.qnx.com/sf/go/post89393 Does io-pkt support anything comparable to packet_mmap in Linux? Apparently it is a zero-copy capture scheme, useful for snffers. Exists in Linux kernels 2.4, 2.6 and.2.8. Description here: http://lingrok.org/source/xref/linux-2.6-linus/Documentation/networking/packet_mmap.txt Dennis Kellly 2011-10-14T10:54:41Z http://community.qnx.com/sf/go/post89319 We're endeavoring to migrate our product from 6.3.x and 6.5.0 (I know, we're late to the party) and we're having trouble getting our io-net 'filter' ported. I put 'filter' in quotes because it's not really a filter, it's more of a tap to allow insertion and extraction of proprietary protocol packets over Ethernet. We initially headed down the BPF route. The BPF migration was easy and seemed to work until we tried running on some of our old hardware with the system fully loaded. Things didn't go so well. The data we're sending is time critical control information and I recently saw that BPF is non-deterministic (post22779). That's not good for us. So, our alternative seems to be lsm-nraw. I got a binary from our FAE and tried it with the pktdump sample but that didn't go well either. Perhaps I'm missing something. I mounted lsm-nraw.so and it seemed to work, I see the /dev/socket/nraw/wm0 entry. When I started pktdump, I was expecting to see a bunch of data but got nothing. I tried pinging the device using the interface with nraw attached but saw nothing from pktdump. What am I missing? I tried to rebuild lsm-nraw.so but I seem to be missing a bunch of header files. Would one of the QNX folks attach an x86 lsm-nraw.so build that I could try. Any other ideas about why the pktdump sample isn't working would be appreciated. I'm reluctant to port our protocol code until I believe I can at least see packets using lsm-nraw. Thanks. Mark Mark Dowdy 2011-10-12T00:12:18Z http://community.qnx.com/sf/go/post89306 I was recently granted access to the networking project. I was hoping to find source for tcpdump/pcap to discover how to set the network adapter into promiscuous mode. However, when following the link from the source guide to the network repository or attempting to use svn, I keep getting an access/permission denied response. Could an admin help me with this? Thanks, Charles Charles Snyder 2011-10-11T16:06:56Z http://community.qnx.com/sf/go/post89281 OK Great! That means it should be OK to use the devn-micrel8841.so driver from QNX 6.4.0 on QNX 6.3.0. Kushal Koolwal 2011-10-10T19:07:44Z http://community.qnx.com/sf/go/post89280 No. You can use devn-* and devnp-* within io-pkt. Only devn-* can be used in io-net. Regards, -seanb ----- Original Message ----- From: Kushal Koolwal [mailto:community-noreply@qnx.com] Sent: Monday, October 10, 2011 12:39 PM To: general-networking <post89279@community.qnx.com> Subject: Using io-pkt networking drivers with io-net Can we use a networking driver which was introduced (with io-pkt stack) in QNX 6.4.0 with QNX 6.3.0 (io-net stack? _______________________________________________ General http://community.qnx.com/sf/go/post89279 Sean Boudreau 2011-10-10T18:43:18Z http://community.qnx.com/sf/go/post89279 Can we use a networking driver which was introduced (with io-pkt stack) in QNX 6.4.0 with QNX 6.3.0 (io-net stack? Kushal Koolwal 2011-10-10T16:39:43Z http://community.qnx.com/sf/go/post88805 Consider a QNX target having a serial port and a PC(using windows) which is connected to QNX target using TCP/IP or UDP. How can a program working on windows control the serial port on the QNX target? Is this related to virtual ports on windows. What should I do? Serkan Öztürk 2011-09-13T20:48:15Z http://community.qnx.com/sf/go/post88752 On Thu, Sep 08, 2011 at 06:52:29PM -0400, sofia escobar wrote: > Is it normal that the IPv6 addres i get from ifconfig is different than the one i get when i use the system call getifaddrs()??? > > i've seen this behavior also in QNX Neutrino 6.4 Answered in general-community. Regards. -seanb Sean Boudreau 2011-09-09T12:49:28Z http://community.qnx.com/sf/go/post88736 Is it normal that the IPv6 addres i get from ifconfig is different than the one i get when i use the system call getifaddrs()??? i've seen this behavior also in QNX Neutrino 6.4 sofia escobar 2011-09-08T22:52:28Z http://community.qnx.com/sf/go/post88230 Hi group, I am working for a network device driver in QNX. I need your help in understanding DMA concepts. Could you please share me some samples for 'how to use DMA in Tx and Rx paths in driver" Is there any dma_map_single() dma_unmap_single() kind of functions in QNX, if not how to achieve such functionality in QNX. Please help me in understanding the DMA operations in QNX. 1. How to allocate DMA SAFE memory in QNX? 2. How to map a memory for DMA_TO_DEVICE or DMA_FROM_DEVICE. 3. How to change the dma memory ownership between CPU and DEVICE. 4. Do we need a typed memory for my network driver dma memory? If so how to do that? 5. Where can I find the complete DMA operations help in QNX? Thanks, Ravinder Are ravinder are 2011-08-22T05:51:03Z http://community.qnx.com/sf/go/post88118 A good reason to not try this would be the underlying bus architecture. SPI is a pure master / slave bus, and the slaves are not able to initate any actions on their own behalf. Modeling a peer-to-peer behavior would require the master to (very) frequently poll for available data from the slave. That incurs overhead on the master side, and also potentially long delays (worst case at least a ticksize) for outgoing frames from the slave. So unless the slave device has some means to signal an interrupt to the master CPU, 'upstream' performance may be poor. Of course the slave could buffer a large number of frames to transmit at once, so the effective throughput might not even be so bad, but if you tried to ping-pong, things'd get slooow. - Thomas Thomas Haupt 2011-08-16T14:47:31Z http://community.qnx.com/sf/go/post87671 Hi Sean Boudreau, We have mounted nfm-bpf.so and nfm-nraw.so in QNX 6.3.2 I have learned bpf can be used to receive and nraw to send raw packets in QNX6.3.2. We are successful in testing receive function. How can we use nraw to send raw packets? Can I get any documents or sample code to send raw packets using "nraw"? Please help. Nagarjuna b 2011-07-29T06:02:30Z http://community.qnx.com/sf/go/post87507 Sean Boudreau Thankyou very much for your reply :) Nagarjuna b 2011-07-21T16:57:17Z http://community.qnx.com/sf/go/post87506 BPF was read only in 6.3.2. Write wasn't implemented. Regards, -seanb ----- Original Message ----- From: Nagarjuna b [mailto:community-noreply@qnx.com] Sent: Thursday, July 21, 2011 12:17 PM To: general-networking <post87505@community.qnx.com> Subject: Re: RE: RE: Send Raw ethernet packet using bpf hi Sean Boudreau, Thankyou for quick reply. Don't we have any option to implement in QNX 6.3.2 ? As of now there is no scope for me to upgrading to higher version of QNX . Also please let me know why am i getting that error? "write:function not implemenmted" _______________________________________________ General http://community.qnx.com/sf/go/post87505 Sean Boudreau 2011-07-21T16:23:50Z http://community.qnx.com/sf/go/post87505 hi Sean Boudreau, Thankyou for quick reply. Don't we have any option to implement in QNX 6.3.2 ? As of now there is no scope for me to upgrading to higher version of QNX . Also please let me know why am i getting that error? "write:function not implemenmted" Nagarjuna b 2011-07-21T16:17:29Z http://community.qnx.com/sf/go/post87504 Sorry!! The previous post was not complete. I Tried the code posted above (icmp.c) Compilation was success. When i run it says write: function not implemented. Please help. Nagarjuna b 2011-07-21T16:11:43Z http://community.qnx.com/sf/go/post87503 bpf wasn't officially supported in 6.3.2. It was read only. I suggest going to 6.5 is possible. Regards, -seanb On Thu, Jul 21, 2011 at 12:07:54PM -0400, Nagarjuna b wrote: > Hi Robert Craig, > > I am also trying to send/receive raw packets using BPF. I am using QNX 6.3.2. On QNX 6.3.2 BPF was not there. > > I mounted nfm-bpf.so using ???mount ???T io-net nfm-bpf.so??? command. > > On mounting ???en-en0??? was created in /dev/io-net. And also bpf is available in /dev/socket in QNX 6.3.2. > > I did a sample program to open bpf . It was successful. > When i am running using "./icmp interface src-mac src-ip dst-mac dst-ip" > its saying > icmp: write: Function not implemented. > > I am new to QNX and BPF. No idea about the error. > May i know why I am facing this error. Thanks in advance. > > > > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post87502 Sean Boudreau 2011-07-21T16:09:41Z http://community.qnx.com/sf/go/post87502 Hi Robert Craig, I am also trying to send/receive raw packets using BPF. I am using QNX 6.3.2. On QNX 6.3.2 BPF was not there. I mounted nfm-bpf.so using “mount –T io-net nfm-bpf.so” command. On mounting “en-en0” was created in /dev/io-net. And also bpf is available in /dev/socket in QNX 6.3.2. I did a sample program to open bpf . It was successful. When i am running using "./icmp interface src-mac src-ip dst-mac dst-ip" its saying icmp: write: Function not implemented. I am new to QNX and BPF. No idea about the error. May i know why I am facing this error. Thanks in advance. Nagarjuna b 2011-07-21T16:07:53Z http://community.qnx.com/sf/go/post87028 Hi, use data exchange over shared memory synchronized by in-memory semaphores ... or port the Binder module of Android to QNX6. Regards --Armin http://www.steinhoff-automation.com Simon Platten wrote: > Hello, > > I am working on an application which involves two processes. The first process (A) acquires data from hardware at high speed, this data is buffered by the process. This process does nothing with the data except gather it up ready to deliver to the main application process. > > The main application process (B) then requests / downloads the data from the buffers. I need to implement a robust communication method between the two processes, where the application process (B) can request data from process (A) reliably and without interrupting (A). > > One of the problems I'm having is that process (A) waits for an interrupt to trigger a thread, the thread is running at high priority waiting for the interrupt to inform it there is some data read to acquire. Once that happens the process populates the ring buffer. However, whilst this is going on the thread dominates and process (B) cannot get any access until the download is finished. > > What can I do to allow process (B) to get access without making process (A) less reliable? > > Thank you. > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post87026 > > Armin Steinhoff 2011-07-04T08:28:48Z http://community.qnx.com/sf/go/post87026 Hello, I am working on an application which involves two processes. The first process (A) acquires data from hardware at high speed, this data is buffered by the process. This process does nothing with the data except gather it up ready to deliver to the main application process. The main application process (B) then requests / downloads the data from the buffers. I need to implement a robust communication method between the two processes, where the application process (B) can request data from process (A) reliably and without interrupting (A). One of the problems I'm having is that process (A) waits for an interrupt to trigger a thread, the thread is running at high priority waiting for the interrupt to inform it there is some data read to acquire. Once that happens the process populates the ring buffer. However, whilst this is going on the thread dominates and process (B) cannot get any access until the download is finished. What can I do to allow process (B) to get access without making process (A) less reliable? Thank you. Simon Platten 2011-07-04T07:48:53Z http://community.qnx.com/sf/go/post86959 Thanks for this info ! It helps a lot! I ask access to the source code of the networking project on Foundry27 a few weeks ago, I'm now registred as part of the project but I can access them... Can you do me a favor and send me the sources of netstat implementation for QNX... I'm struggling in trying to do the same than the FreeBSD sources with kvm... Cheers, Daniel -----Original Message----- From: Patrik Lahti [mailto:community-noreply@qnx.com] Sent: mercredi 29 juin 2011 19:18 To: general-networking Subject: Re: sysctl net.inet.tcp.pcblist in our software # sysctl net.inet.tcp.pcblist sysctl: net.inet.tcp.pcblist: use 'netstat' or 'sockstat' to view this information I believe netstat in turn uses the kvm interface, which is in kvm.h. Regards, /P On 06/29/2011 01:11 PM, Daniel Peppicelli wrote: > > Hi everyone ! > > I'm facing an issue in one program I'm currently writting. I need to > get a list of all opened tcp sockets, their status and the local and > remote port to where they are bind. > For this, I was thinking of using netbsd sysctl functionality. (We do > not want to have netstat on our embedded system). This is the piece of > code I did write to test it out: > > int mib[4]; > size_t namelen = 0; > > char temp[1000]; > size_t len = sizeof(temp); > > if (-1 == sysctlnametomib("net.inet.tcp.pcblist", mib, &namelen)) > { > LOG_SYSTEM_DEBUG("error: %d",errno); > } > if (-1 == sysctl(mib, namelen, &temp, &len, NULL, 0)) > { > LOG_SYSTEM_DEBUG("error: %d",errno); > } > > Unfortunately the second instruction fails and errno is set to 22 > EINVAL. According to QNX documentation, this error happens when: "The > name array is less than two or greater than CTL_MAXNAME entries long, > or a non-NULL newp was given, and its specified length in newlen is > too large or too small.". For me, this is not the case, everythings > seems okay. Even if I replace &temp by NULL in the second call, in > order to have the length required, it fails the same... > > Do you have maybe any idea about this ? > > Many thanks in advance, > > Daniel Peppicelli > > Please be advised that this email may contain confidential > information. If you are not the intended recipient, please notify us > by email by replying to the sender and delete this message. The > sender disclaims that the content of this email constitutes an offer > to enter into, or the acceptance of, any agreement; provided that the > foregoing does not invalidate the binding effect of any digital or > other electronic reproduction of a manual signature that is included > in any attachment. > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post86946 > _______________________________________________ General http://community.qnx.com/sf/go/post86947 Please be advised that this email may contain confidential information. If you are not the intended recipient, please notify us by email by replying to the sender and delete this message. The sender disclaims that the content of this email constitutes an offer to enter into, or the acceptance of, any agreement; provided that the foregoing does not invalidate the binding effect of any digital or other electronic reproduction of a manual signature that is included in any attachment. Daniel Peppicelli 2011-06-30T06:55:43Z http://community.qnx.com/sf/go/post86947 # sysctl net.inet.tcp.pcblist sysctl: net.inet.tcp.pcblist: use 'netstat' or 'sockstat' to view this information I believe netstat in turn uses the kvm interface, which is in kvm.h. Regards, /P On 06/29/2011 01:11 PM, Daniel Peppicelli wrote: > > Hi everyone ! > > I'm facing an issue in one program I'm currently writting. I need to > get a list of all opened tcp sockets, their status and the local and > remote port to where they are bind. > For this, I was thinking of using netbsd sysctl functionality. (We do > not want to have netstat on our embedded system). This is the piece of > code I did write to test it out: > > int mib[4]; > size_t namelen = 0; > > char temp[1000]; > size_t len = sizeof(temp); > > if (-1 == sysctlnametomib("net.inet.tcp.pcblist", mib, &namelen)) > { > LOG_SYSTEM_DEBUG("error: %d",errno); > } > if (-1 == sysctl(mib, namelen, &temp, &len, NULL, 0)) > { > LOG_SYSTEM_DEBUG("error: %d",errno); > } > > Unfortunately the second instruction fails and errno is set to 22 > EINVAL. According to QNX documentation, this error happens when: "The > name array is less than two or greater than CTL_MAXNAME entries long, > or a non-NULL newp was given, and its specified length in newlen is > too large or too small.". For me, this is not the case, everythings > seems okay. Even if I replace &temp by NULL in the second call, in > order to have the length required, it fails the same... > > Do you have maybe any idea about this ? > > Many thanks in advance, > > Daniel Peppicelli > > Please be advised that this email may contain confidential > information. If you are not the intended recipient, please notify us > by email by replying to the sender and delete this message. The > sender disclaims that the content of this email constitutes an offer > to enter into, or the acceptance of, any agreement; provided that the > foregoing does not invalidate the binding effect of any digital or > other electronic reproduction of a manual signature that is included > in any attachment. > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post86946 > Patrik Lahti 2011-06-29T17:17:30Z http://community.qnx.com/sf/go/post86946 Hi everyone ! I'm facing an issue in one program I'm currently writting. I need to get a list of all opened tcp sockets, their status and the local and remote port to where they are bind. For this, I was thinking of using netbsd sysctl functionality. (We do not want to have netstat on our embedded system). This is the piece of code I did write to test it out: int mib[4]; size_t namelen = 0; char temp[1000]; size_t len = sizeof(temp); if (-1 == sysctlnametomib("net.inet.tcp.pcblist", mib, &namelen)) { LOG_SYSTEM_DEBUG("error: %d",errno); } if (-1 == sysctl(mib, namelen, &temp, &len, NULL, 0)) { LOG_SYSTEM_DEBUG("error: %d",errno); } Unfortunately the second instruction fails and errno is set to 22 EINVAL. According to QNX documentation, this error happens when: "The name array is less than two or greater than CTL_MAXNAME entries long, or a non-NULL newp was given, and its specified length in newlen is too large or too small.". For me, this is not the case, everythings seems okay. Even if I replace &temp by NULL in the second call, in order to have the length required, it fails the same... Do you have maybe any idea about this ? Many thanks in advance, Daniel Peppicelli Please be advised that this email may contain confidential information. If you are not the intended recipient, please notify us by email by replying to the sender and delete this message. The sender disclaims that the content of this email constitutes an offer to enter into, or the acceptance of, any agreement; provided that the foregoing does not invalidate the binding effect of any digital or other electronic reproduction of a manual signature that is included in any attachment. Daniel Peppicelli 2011-06-29T17:10:59Z http://community.qnx.com/sf/go/post86465 I have come across a DHCP server faulty, and do not response for the DHCP discover. when I run DHCP server in debug mode,i got the follwing debug information: DHCPDISCOVER from 00:80:66:04:a0:8f via vlan42 icmp_echorequest 192.168.11.138: Host is down DHCPOFFER on 192.168.11.138 to 00:80:66:04:a0:8f (EIM) via vlan42 Updating ARP cache entry for 192.168.11.138 Current ARP cache entry being used send_packet: Host is down Can anyone help to explain the reason or what the following error message refering to: 1. icmp_echorequest 192.168.11.138: Host is down 2.send_packet: Host is down I am using QNX640 on MPC8313 Ji Dong 2011-06-07T08:53:28Z http://community.qnx.com/sf/go/post85605 Hello, In my host (a x86 based board running on QNX 6.5), there is a need to be able to listen to network packets coming over different multi-cast addresses but the same physical ethernet link. For example, my host should be able to listen to packets coming on multi-cast addresses (224.0.0.1 & 224.0.0.2). To listen to packets on a single multi-cast address, we used the setsockopt() call with IP_ADD_MEMBERSHIP option & gave the appropriate multi-cast address. Now can I make multiple setsockopt() calls for each of the different multi-cast addresses so that my socket is capable of listening to packets on all these multi-cast addresses? For example, here's a sample code: // Begin Sample code struct ip_mreq mreq1; struct ip_mreq mreq2; setsockopt(s, IPPROTO_IP, IP_ADD_MEMBERSHIP, &mreq1, sizeof(mreq1)); setsockopt(s, IPPROTO_IP, IP_ADD_MEMBERSHIP, &mreq2, sizeof(mreq2)); // End of sample code I am hoping that the second setsockopt() call does not over-write the options set using the 1st setsockopt() call. Any help on this will be appreciated. Regards Sunil Keshava Sunil Keshava 2011-05-12T05:43:57Z http://community.qnx.com/sf/go/post85412 Hi, I am new to QNX... When trying to compile the code provide in: http://www.qnx.com/developers/docs/6.5.0/index.jsp?topic=/com.qnx.doc.io-pkt_en_user_guide/filtering.html I got the following error: C:\BBNDK100\host\win32\x86\usr\bin\make.exe -k CPULIST=x86 all --file=C:/DOCUME~1/myin/LOCALS~1/Temp/QMakefile130454651754650171.tmp C:/BBNDK100/host/win32/x86/usr/bin/make.exe -j 1 -Cx86 -fMakefile all make.exe[1]: Entering directory `C:/ide-4.7-workspace/PFTest/x86' C:/BBNDK100/host/win32/x86/usr/bin/make.exe -j 1 -Co -fMakefile all make.exe[2]: Entering directory `C:/ide-4.7-workspace/PFTest/x86/o' C:/BBNDK100/host/win32/x86/usr/bin/qcc -Vgcc_ntox86 -c -Wc,-Wall -O -DNDEBUG -I. -IC:/ide-4.7-workspace/PFTest/x86/o -IC:/ide-4.7-workspace/PFTest/x86 -IC:/ide-4.7-workspace/PFTest -IC:/BBNDK100/target/qnx6/usr/include/io-pkt -IC:/BBNDK100/target/qnx6/usr/include -DBUILDENV_qss C:/ide-4.7-workspace/PFTest/PFTest.c C:/BBNDK100/host/win32/x86/usr/bin/rm -f C:/ide-4.7-workspace/PFTest/x86/o/PFTest C:/BBNDK100/host/win32/x86/usr/bin/qcc -Vgcc_ntox86 -ptcpip -o C:/ide-4.7-workspace/PFTest/x86/o/PFTest PFTest.o -L . -L C:/BBNDK100/target/qnx6/x86/lib -L C:/BBNDK100/target/qnx6/x86/usr/lib -Wl,--rpath-link . -Wl,--rpath-link C:/BBNDK100/target/qnx6/x86/lib -Wl,--rpath-link C:/BBNDK100/target/qnx6/x86/usr/lib -Bstatic -lc -Bdynamic -lc -Bstatic -lsocket -Bdynamic -lsocket -Bstatic -lsocks -Bdynamic -lsocks PFTest.o: In function `pfil_entry': PFTest.c:(.text+0x51): undefined reference to `pfil_head_get' PFTest.c:(.text+0x7b): undefined reference to `pfil_add_hook' PFTest.c:(.text+0x9b): undefined reference to `pfil_add_hook' PFTest.c:(.text+0xaf): undefined reference to `pfil_head_get' PFTest.c:(.text+0xd5): undefined reference to `pfil_add_hook' PFTest.c:(.text+0xf5): undefined reference to `pfil_add_hook' There is an old thread about this issue, but there seems no final solutions: http://community.qnx.com/sf/discussion/do/listPosts/projects.networking/discussion.drivers.topc10823 Did I miss something in Momentics Project Properties settings? Thanks, Matthew Matthew Yin 2011-05-04T22:24:26Z http://community.qnx.com/sf/go/post85409 We have a windows machine using 2 NIC connected to a cisco switch, the idea is somewhat double avaiable bandwidth. With this configuration we have unexpected packet lost occuring in the switch. Our IT guy changed the configuration from IEEE Load Aggreagation to Team ABL (I'm just repeating what I'm told). This has greatly improve the situation, however with TEAM ABL sloginfo is full of May 05 18:11:06 5 14 0 arp info overwritten for 10.121.205.100 by 00:1b:21:9f:0d:96 May 05 18:11:38 5 14 0 arp info overwritten for 10.121.205.100 by 00:1b:21:9f:0d:97 I've always been tough that having two NICs using the same IP adress is really bad. Yet this seems to be working. Does it work by sheer luck or by design? Mario Charest 2011-05-04T18:32:22Z http://community.qnx.com/sf/go/post85320 I'm trying to install Samba 3.5.8 on a QNX 6.5.0 x86 system. Below is my debug file. I know it's a lot, but if someone can help me make sense of it that would be great. [2011/04/29 10:55:39.968657, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 484B4C4D2F5359535445 [2011/04/29 10:55:39.969657, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6fce8 [2011/04/29 10:55:39.969657, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 484B4C4D2F5359535445 [2011/04/29 10:55:39.969657, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 53414D42415F534F5254 [2011/04/29 10:55:39.969657, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6ffe8 [2011/04/29 10:55:39.969657, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 53414D42415F534F5254 [2011/04/29 10:55:39.970657, 5] lib/util_tdb.c:521(tdb_wrap_log) tdb(/usr/local/samba/var/locks/registry.tdb): tdb_transaction_start: nesting 1 [2011/04/29 10:55:39.970657, 5] lib/util_tdb.c:521(tdb_wrap_log) tdb(/usr/local/samba/var/locks/registry.tdb): tdb_transaction_start: nesting 2 [2011/04/29 10:55:39.970657, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 53414D42415F534F5254 [2011/04/29 10:55:39.970657, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6eae0 [2011/04/29 10:55:39.970657, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 53414D42415F534F5254 [2011/04/29 10:55:39.971656, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services/WINS] [2011/04/29 10:55:40.425587, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [HKLM] [2011/04/29 10:55:40.425587, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (4) [2011/04/29 10:55:40.425587, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM] [2011/04/29 10:55:40.425587, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM] [2011/04/29 10:55:40.425587, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:40.425587, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM] [2011/04/29 10:55:40.426587, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM] [2011/04/29 10:55:40.426587, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [SYSTEM] [2011/04/29 10:55:40.426587, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (5) [2011/04/29 10:55:40.426587, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM] [2011/04/29 10:55:40.426587, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM] [2011/04/29 10:55:40.426587, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:40.426587, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM] [2011/04/29 10:55:40.427587, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM] [2011/04/29 10:55:40.427587, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [CurrentControlSet] [2011/04/29 10:55:40.427587, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:40.427587, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet] [2011/04/29 10:55:40.427587, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet] [2011/04/29 10:55:40.427587, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:40.427587, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet] [2011/04/29 10:55:40.428587, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet] [2011/04/29 10:55:40.428587, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:40.428587, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [Services] [2011/04/29 10:55:40.428587, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:40.428587, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet/Services] [2011/04/29 10:55:40.428587, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet/Services] [2011/04/29 10:55:40.428587, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:40.429586, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet/Services] [2011/04/29 10:55:40.429586, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet\Services] [2011/04/29 10:55:40.429586, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:40.429586, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [WINS] [2011/04/29 10:55:40.429586, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:40.429586, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet/Services/WINS] [2011/04/29 10:55:40.430586, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet/Services/WINS] [2011/04/29 10:55:40.430586, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:40.430586, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet/Services/WINS] [2011/04/29 10:55:40.430586, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:40.430586, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:40.430586, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:40.430586, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (5) [2011/04/29 10:55:40.431586, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:40.431586, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:40.431586, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:40.432586, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 484B4C4D2F5359535445 [2011/04/29 10:55:40.432586, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a5f978 [2011/04/29 10:55:40.448583, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 484B4C4D2F5359535445 [2011/04/29 10:55:40.448583, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 53414D42415F534F5254 [2011/04/29 10:55:40.448583, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6eae0 [2011/04/29 10:55:40.448583, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 53414D42415F534F5254 [2011/04/29 10:55:40.448583, 5] lib/util_tdb.c:521(tdb_wrap_log) tdb(/usr/local/samba/var/locks/registry.tdb): tdb_transaction_start: nesting 1 [2011/04/29 10:55:40.449583, 5] lib/util_tdb.c:521(tdb_wrap_log) tdb(/usr/local/samba/var/locks/registry.tdb): tdb_transaction_start: nesting 2 [2011/04/29 10:55:40.449583, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 53414D42415F534F5254 [2011/04/29 10:55:40.449583, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6d258 [2011/04/29 10:55:40.449583, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 53414D42415F534F5254 [2011/04/29 10:55:40.449583, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services\WINS/Security] [2011/04/29 10:55:40.857521, 10] registry/reg_backend_db.c:1663(regdb_store_values_internal) regdb_store_values: Looking for value of key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:40.858521, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 53414D42415F52454756 [2011/04/29 10:55:40.858521, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6d258 [2011/04/29 10:55:40.858521, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 53414D42415F52454756 [2011/04/29 10:55:41.282456, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (4) [2011/04/29 10:55:41.282456, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [HKLM] [2011/04/29 10:55:41.282456, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (4) [2011/04/29 10:55:41.282456, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM] [2011/04/29 10:55:41.282456, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM] [2011/04/29 10:55:41.282456, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:41.283456, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM] [2011/04/29 10:55:41.283456, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM] [2011/04/29 10:55:41.283456, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [SYSTEM] [2011/04/29 10:55:41.283456, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (5) [2011/04/29 10:55:41.283456, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM] [2011/04/29 10:55:41.283456, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM] [2011/04/29 10:55:41.284456, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:41.284456, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM] [2011/04/29 10:55:41.284456, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM] [2011/04/29 10:55:41.284456, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [CurrentControlSet] [2011/04/29 10:55:41.284456, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:41.284456, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet] [2011/04/29 10:55:41.284456, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet] [2011/04/29 10:55:41.285455, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:41.285455, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet] [2011/04/29 10:55:41.285455, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet] [2011/04/29 10:55:41.285455, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:41.285455, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [Services] [2011/04/29 10:55:41.285455, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:41.285455, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet/Services] [2011/04/29 10:55:41.286455, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet/Services] [2011/04/29 10:55:41.286455, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:41.286455, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet/Services] [2011/04/29 10:55:41.286455, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet\Services] [2011/04/29 10:55:41.286455, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:41.287455, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [WINS] [2011/04/29 10:55:41.287455, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:41.287455, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet/Services/WINS] [2011/04/29 10:55:41.287455, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet/Services/WINS] [2011/04/29 10:55:41.287455, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:41.287455, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet/Services/WINS] [2011/04/29 10:55:41.287455, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet\Services\WINS] [2011/04/29 10:55:41.288455, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:41.288455, 7] registry/reg_api.c:133(regkey_open_onelevel) regkey_open_onelevel: name = [Security] [2011/04/29 10:55:41.288455, 10] registry/reg_backend_db.c:451(regdb_open) regdb_open: incrementing refcount (6) [2011/04/29 10:55:41.288455, 10] registry/reg_cachehook.c:125(reghook_cache_find) reghook_cache_find: Searching for keyname [/HKLM/SYSTEM/CurrentControlSet/Services/WINS/Security] [2011/04/29 10:55:41.288455, 10] lib/adt_tree.c:352(pathtree_find) pathtree_find: Enter [/HKLM/SYSTEM/CurrentControlSet/Services/WINS/Security] [2011/04/29 10:55:41.288455, 10] lib/adt_tree.c:425(pathtree_find) pathtree_find: Exit [2011/04/29 10:55:41.288455, 10] registry/reg_cachehook.c:130(reghook_cache_find) reghook_cache_find: found ops 8a55ea0 for key [/HKLM/SYSTEM/CurrentControlSet/Services/WINS/Security] [2011/04/29 10:55:41.288455, 10] registry/reg_backend_db.c:1483(regdb_fetch_keys_internal) regdb_fetch_keys: no subkeys found for key [HKLM\SYSTEM\CurrentControlSet\Services\WINS\Security] [2011/04/29 10:55:41.289455, 10] registry/reg_backend_db.c:1726(regdb_get_secdesc) regdb_get_secdesc: Getting secdesc of key [HKLM\SYSTEM\CurrentControlSet\Services\WINS\Security] [2011/04/29 10:55:41.289455, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (6) [2011/04/29 10:55:41.289455, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (5) [2011/04/29 10:55:41.289455, 10] registry/reg_backend_db.c:1663(regdb_store_values_internal) regdb_store_values: Looking for value of key [HKLM\SYSTEM\CurrentControlSet\Services\WINS\Security] [2011/04/29 10:55:41.289455, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 53414D42415F52454756 [2011/04/29 10:55:41.290455, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6eae0 [2011/04/29 10:55:41.290455, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 53414D42415F52454756 [2011/04/29 10:55:41.798377, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (4) [2011/04/29 10:55:41.798377, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (3) [2011/04/29 10:55:41.831372, 10] registry/reg_backend_db.c:485(regdb_close) regdb_close: decrementing refcount (2) [2011/04/29 10:55:41.860367, 10] printing/nt_printing.c:561(nt_printing_init) Fresh database [2011/04/29 10:55:41.862367, 10] printing/nt_printing.c:707(update_c_setprinter) update_c_setprinter: c_setprinter = 0 [2011/04/29 10:55:41.863367, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.863367, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.863367, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.863367, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.863367, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.863367, 6] passdb/pdb_interface.c:285(pdb_getsampwsid) pdb_getsampwsid: Building guest account [2011/04/29 10:55:41.864367, 5] lib/username.c:133(Get_Pwnam_alloc) Finding user dfento01 [2011/04/29 10:55:41.864367, 5] lib/username.c:77(Get_Pwnam_internals) Trying _Get_Pwnam(), username as lowercase is dfento01 [2011/04/29 10:55:41.864367, 5] lib/username.c:110(Get_Pwnam_internals) Get_Pwnam_internals did find user [dfento01]! [2011/04/29 10:55:41.864367, 10] passdb/pdb_get_set.c:608(pdb_set_username) pdb_set_username: setting username dfento01, was [2011/04/29 10:55:41.864367, 10] passdb/pdb_get_set.c:677(pdb_set_fullname) pdb_set_full_name: setting full name Dan Fenton, was [2011/04/29 10:55:41.864367, 10] passdb/pdb_get_set.c:631(pdb_set_domain) pdb_set_domain: setting domain QNX65_NHP, was [2011/04/29 10:55:41.864367, 10] passdb/pdb_get_set.c:537(pdb_set_user_sid) pdb_set_user_sid: setting user sid S-1-5-21-2339009583-1887513364-3383096932-501 [2011/04/29 10:55:41.865367, 10] passdb/pdb_compat.c:72(pdb_set_user_sid_from_rid) pdb_set_user_sid_from_rid: setting user sid S-1-5-21-2339009583-1887513364-3383096932-501 from rid 501 [2011/04/29 10:55:41.865367, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.865367, 5] lib/username.c:133(Get_Pwnam_alloc) Finding user dfento01 [2011/04/29 10:55:41.865367, 5] lib/username.c:77(Get_Pwnam_internals) Trying _Get_Pwnam(), username as lowercase is dfento01 [2011/04/29 10:55:41.865367, 5] lib/username.c:110(Get_Pwnam_internals) Get_Pwnam_internals did find user [dfento01]! [2011/04/29 10:55:41.865367, 5] lib/username.c:133(Get_Pwnam_alloc) Finding user dfento01 [2011/04/29 10:55:41.865367, 5] lib/username.c:77(Get_Pwnam_internals) Trying _Get_Pwnam(), username as lowercase is dfento01 [2011/04/29 10:55:41.865367, 5] lib/username.c:110(Get_Pwnam_internals) Get_Pwnam_internals did find user [dfento01]! [2011/04/29 10:55:41.865367, 10] lib/system_smbd.c:122(sys_getgrouplist) sys_getgrouplist: user [dfento01] [2011/04/29 10:55:41.920358, 10] lib/gencache.c:334(gencache_get_data_blob) Cache entry with key = IDMAP/GID2SID/4 couldn't be found [2011/04/29 10:55:41.921358, 5] passdb/lookup_sid.c:1387(gid_to_sid) gid_to_sid: winbind failed to find a sid for gid 4 [2011/04/29 10:55:41.921358, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.921358, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.921358, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.921358, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.921358, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.921358, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.922358, 10] passdb/lookup_sid.c:1182(legacy_gid_to_sid) LEGACY: gid 4 -> sid S-1-22-2-4 [2011/04/29 10:55:41.922358, 5] auth/auth_util.c:649(make_server_info_sam) make_server_info_sam: made server info for user dfento01 -> dfento01 [2011/04/29 10:55:41.922358, 10] auth/token_util.c:356(create_local_nt_token) Create local NT token for S-1-5-21-2339009583-1887513364-3383096932-501 [2011/04/29 10:55:41.922358, 10] lib/gencache.c:334(gencache_get_data_blob) Cache entry with key = IDMAP/SID2GID/S-1-5-32-544 couldn't be found [2011/04/29 10:55:41.922358, 10] passdb/lookup_sid.c:1511(sid_to_gid) winbind failed to find a gid for sid S-1-5-32-544 [2011/04/29 10:55:41.922358, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.922358, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.922358, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.922358, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.923358, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.923358, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.923358, 10] passdb/lookup_sid.c:1256(legacy_sid_to_gid) LEGACY: mapping failed for sid S-1-5-32-544 [2011/04/29 10:55:41.923358, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.923358, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.923358, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.923358, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.924358, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.924358, 3] auth/token_util.c:436(create_local_nt_token) Failed to fetch domain sid for QNXHARRIS.NET [2011/04/29 10:55:41.924358, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.924358, 10] lib/gencache.c:334(gencache_get_data_blob) Cache entry with key = IDMAP/SID2GID/S-1-5-32-545 couldn't be found [2011/04/29 10:55:41.924358, 10] passdb/lookup_sid.c:1511(sid_to_gid) winbind failed to find a gid for sid S-1-5-32-545 [2011/04/29 10:55:41.924358, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.924358, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.924358, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.925357, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.925357, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.925357, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.925357, 10] passdb/lookup_sid.c:1256(legacy_sid_to_gid) LEGACY: mapping failed for sid S-1-5-32-545 [2011/04/29 10:55:41.925357, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.925357, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.925357, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.926357, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.926357, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.926357, 3] auth/token_util.c:467(create_local_nt_token) Failed to fetch domain sid for QNXHARRIS.NET [2011/04/29 10:55:41.926357, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.926357, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.926357, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.926357, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.926357, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.926357, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.928357, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.928357, 3] lib/privileges.c:63(get_privileges) get_privileges: No privileges assigned to SID [S-1-5-21-2339009583-1887513364-3383096932-501] [2011/04/29 10:55:41.928357, 3] lib/privileges.c:63(get_privileges) get_privileges: No privileges assigned to SID [S-1-22-2-4] [2011/04/29 10:55:41.928357, 5] lib/privileges.c:128(get_privileges_for_sids) get_privileges_for_sids: sid = S-1-1-0 Privilege set: SE_PRIV 0x0 0x0 0x0 0x0 [2011/04/29 10:55:41.928357, 3] lib/privileges.c:63(get_privileges) get_privileges: No privileges assigned to SID [S-1-5-2] [2011/04/29 10:55:41.929357, 3] lib/privileges.c:63(get_privileges) get_privileges: No privileges assigned to SID [S-1-5-32-546] [2011/04/29 10:55:41.937356, 10] lib/gencache.c:334(gencache_get_data_blob) Cache entry with key = IDMAP/SID2GID/S-1-1-0 couldn't be found [2011/04/29 10:55:41.937356, 10] passdb/lookup_sid.c:1511(sid_to_gid) winbind failed to find a gid for sid S-1-1-0 [2011/04/29 10:55:41.937356, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.938355, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.938355, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.938355, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.938355, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.938355, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.938355, 10] passdb/lookup_sid.c:1256(legacy_sid_to_gid) LEGACY: mapping failed for sid S-1-1-0 [2011/04/29 10:55:41.938355, 10] auth/auth_util.c:753(create_local_token) Could not convert SID S-1-1-0 to gid, ignoring it [2011/04/29 10:55:41.939355, 10] lib/gencache.c:334(gencache_get_data_blob) Cache entry with key = IDMAP/SID2GID/S-1-5-2 couldn't be found [2011/04/29 10:55:41.939355, 10] passdb/lookup_sid.c:1511(sid_to_gid) winbind failed to find a gid for sid S-1-5-2 [2011/04/29 10:55:41.939355, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.939355, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.939355, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.939355, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.939355, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.940355, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.940355, 10] passdb/lookup_sid.c:1256(legacy_sid_to_gid) LEGACY: mapping failed for sid S-1-5-2 [2011/04/29 10:55:41.940355, 10] auth/auth_util.c:753(create_local_token) Could not convert SID S-1-5-2 to gid, ignoring it [2011/04/29 10:55:41.940355, 10] lib/gencache.c:334(gencache_get_data_blob) Cache entry with key = IDMAP/SID2GID/S-1-5-32-546 couldn't be found [2011/04/29 10:55:41.940355, 10] passdb/lookup_sid.c:1511(sid_to_gid) winbind failed to find a gid for sid S-1-5-32-546 [2011/04/29 10:55:41.940355, 3] smbd/sec_ctx.c:210(push_sec_ctx) push_sec_ctx(0, 0) : sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.940355, 3] smbd/uid.c:429(push_conn_ctx) push_conn_ctx(0) : conn_ctx_stack_ndx = 0 [2011/04/29 10:55:41.940355, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 1 [2011/04/29 10:55:41.940355, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:55:41.941355, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:55:41.941355, 3] smbd/sec_ctx.c:418(pop_sec_ctx) pop_sec_ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:55:41.941355, 10] passdb/lookup_sid.c:1256(legacy_sid_to_gid) LEGACY: mapping failed for sid S-1-5-32-546 [2011/04/29 10:55:41.941355, 10] auth/auth_util.c:753(create_local_token) Could not convert SID S-1-5-32-546 to gid, ignoring it [2011/04/29 10:55:41.941355, 10] auth/token_util.c:531(debug_nt_user_token) NT user token of user S-1-5-21-2339009583-1887513364-3383096932-501 contains 6 SIDs SID[ 0]: S-1-5-21-2339009583-1887513364-3383096932-501 SID[ 1]: S-1-22-2-4 SID[ 2]: S-1-1-0 SID[ 3]: S-1-5-2 SID[ 4]: S-1-5-32-546 SID[ 5]: S-1-22-1-104 SE_PRIV 0x0 0x0 0x0 0x0 [2011/04/29 10:55:41.942355, 10] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 104 Primary group is 4 and contains 1 supplementary groups Group[ 0]: 4 [2011/04/29 10:55:41.942355, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 494E464F2F7665727369 [2011/04/29 10:55:41.942355, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8af0688 [2011/04/29 10:55:41.943355, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 494E464F2F7665727369 [2011/04/29 10:55:42.821220, 3] printing/printing.c:1412(start_background_queue) start_background_queue: Starting background LPQ thread [2011/04/29 10:55:42.857215, 5] printing/printing.c:1434(start_background_queue) [2011/04/29 10:55:42.857215, 0] lib/util_sock.c:844(open_socket_in) start_background_queue: background LPQ thread started open_socket_in(): socket() call failed: Address family not supported by protocol family [2011/04/29 10:55:42.857215, 0] smbd/server.c:500(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address family not supported by protocol family [2011/04/29 10:55:42.857215, 0] lib/util_sock.c:844(open_socket_in) open_socket_in(): socket() call failed: Address family not supported by protocol family [2011/04/29 10:55:42.857215, 0] smbd/server.c:500(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address family not supported by protocol family [2011/04/29 10:55:42.858215, 10] lib/util_sock.c:893(open_socket_in) bind succeeded on port 445 [2011/04/29 10:55:42.858215, 5] lib/util_sock.c:304(print_socket_options) Socket options: SO_KEEPALIVE = 8 [2011/04/29 10:55:42.858215, 5] smbd/connection.c:142(claim_connection) SO_REUSEADDR = 4 claiming [smbd lpq backend] [2011/04/29 10:55:42.859215, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) SO_BROADCAST = 0 Locking key 17303511FFFFFFFF736D TCP_NODELAY = 0 [2011/04/29 10:55:42.859215, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) IPTOS_LOWDELAY = 0 Allocated locked data 0x8adc780 IPTOS_THROUGHPUT = 0 SO_REUSEPORT = 512 SO_SNDBUF = 22528 SO_RCVBUF = 32768 SO_SNDLOWAT = 18432 SO_RCVLOWAT = 1 SO_SNDTIMEO = 0 SO_RCVTIMEO = 0 [2011/04/29 10:55:42.865214, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) [2011/04/29 10:55:42.865214, 5] lib/util_sock.c:304(print_socket_options) Socket options: SO_KEEPALIVE = 8 SO_REUSEADDR = 4 SO_BROADCAST = 0 TCP_NODELAY = 4 IPTOS_LOWDELAY = 0 IPTOS_THROUGHPUT = 0 SO_REUSEPORT = 512 SO_SNDBUF = 22528 Unlocking key 17303511FFFFFFFF736D SO_RCVBUF = 32768 SO_SNDLOWAT = 18432 [2011/04/29 10:55:42.866213, 5] printing/printing.c:1468(start_background_queue) SO_RCVLOWAT = 1 start_background_queue: background LPQ thread waiting for messages SO_SNDTIMEO = 0 SO_RCVTIMEO = 0 [2011/04/29 10:55:42.866213, 10] lib/util_sock.c:893(open_socket_in) bind succeeded on port 139 [2011/04/29 10:55:42.866213, 5] lib/util_sock.c:304(print_socket_options) Socket options: SO_KEEPALIVE = 8 SO_REUSEADDR = 4 SO_BROADCAST = 0 TCP_NODELAY = 0 IPTOS_LOWDELAY = 0 IPTOS_THROUGHPUT = 0 SO_REUSEPORT = 512 SO_SNDBUF = 22528 SO_RCVBUF = 32768 SO_SNDLOWAT = 18432 SO_RCVLOWAT = 1 SO_SNDTIMEO = 0 SO_RCVTIMEO = 0 [2011/04/29 10:55:42.867213, 5] lib/util_sock.c:304(print_socket_options) Socket options: SO_KEEPALIVE = 8 SO_REUSEADDR = 4 SO_BROADCAST = 0 TCP_NODELAY = 4 IPTOS_LOWDELAY = 0 IPTOS_THROUGHPUT = 0 SO_REUSEPORT = 512 SO_SNDBUF = 22528 SO_RCVBUF = 32768 SO_SNDLOWAT = 18432 SO_RCVLOWAT = 1 SO_SNDTIMEO = 0 SO_RCVTIMEO = 0 [2011/04/29 10:55:42.868213, 5] smbd/connection.c:142(claim_connection) claiming [] [2011/04/29 10:55:42.868213, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 28203511FFFFFFFF0000 [2011/04/29 10:55:42.868213, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8adc780 [2011/04/29 10:55:42.869213, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 28203511FFFFFFFF0000 [2011/04/29 10:55:42.869213, 5] lib/messages.c:297(messaging_register) Overriding messaging pointer for type 1 - private_data=0 [2011/04/29 10:55:42.869213, 2] smbd/server.c:721(smbd_parent_loop) waiting for connections [2011/04/29 10:56:06.977524, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:56:06.977524, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:56:06.978524, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:56:06.978524, 5] smbd/uid.c:369(change_to_root_user) change_to_root_user: now uid=(0,0) gid=(0,0) [2011/04/29 10:56:06.978524, 3] smbd/connection.c:31(yield_connection) Yielding connection to [2011/04/29 10:56:06.978524, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 17303511FFFFFFFF0000 [2011/04/29 10:56:06.979524, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8adc780 [2011/04/29 10:56:06.979524, 3] smbd/connection.c:42(yield_connection) deleting connection record returned NT_STATUS_NOT_FOUND [2011/04/29 10:56:06.979524, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 17303511FFFFFFFF0000 [2011/04/29 10:56:06.979524, 3] smbd/server.c:902(exit_server_common) Server exit (termination signal) [2011/04/29 10:56:07.580432, 3] smbd/sec_ctx.c:310(set_sec_ctx) setting sec ctx (0, 0) - sec_ctx_stack_ndx = 0 [2011/04/29 10:56:07.590430, 5] auth/token_util.c:525(debug_nt_user_token) NT user token: (NULL) [2011/04/29 10:56:07.590430, 5] auth/token_util.c:551(debug_unix_user_token) UNIX token of user 0 Primary group is 0 and contains 0 supplementary groups [2011/04/29 10:56:07.590430, 5] smbd/uid.c:369(change_to_root_user) change_to_root_user: now uid=(0,0) gid=(0,0) [2011/04/29 10:56:07.590430, 3] smbd/connection.c:31(yield_connection) Yielding connection to [2011/04/29 10:56:07.591430, 10] lib/dbwrap_tdb.c:100(db_tdb_fetch_locked) Locking key 28203511FFFFFFFF0000 [2011/04/29 10:56:07.591430, 10] lib/dbwrap_tdb.c:129(db_tdb_fetch_locked) Allocated locked data 0x8a6c870 [2011/04/29 10:56:07.598429, 10] lib/dbwrap_tdb.c:42(db_tdb_record_destr) Unlocking key 28203511FFFFFFFF0000 [2011/04/29 10:56:07.599429, 3] smbd/server.c:902(exit_server_common) Server exit (termination signal) Thank You for your help in advance, -Dan Dan Fenton 2011-04-29T15:57:59Z http://community.qnx.com/sf/go/post85306 Disregard my last post. I'm still having the same issue as stated in my original post. -Dan Dan Fenton 2011-04-28T19:39:40Z http://community.qnx.com/sf/go/post85305 On Thu, Apr 28, 2011 at 02:56:49PM -0400, Dan Fenton wrote: > Wound up just starting form the beginning and re-configuring, > re-compiling, re-making, and re-make install and got it to run and allow > users to connect. Now my problem is when more than one user connects at > a time Samba seems to lock up. Checking the log files for the machines > that were connected gave the following debug message: > > smbd/connection.c:claim_connection(208) > claim_connection: tdb_store failed with error Corrupt database. > > do i need separate lock files for each share? Try commenting out 'HAVE_MMAP' in your config.h after configure but before make. > > can anyone shed some light on this? And I should add again that there > was a problem with -lpthreads so in the make process the Makefile was > edited to not use -lpthreads That shouldn't be a problem. Regards, -seanb Sean Boudreau 2011-04-28T19:32:49Z http://community.qnx.com/sf/go/post85304 Wound up just starting form the beginning and re-configuring, re-compiling, re-making, and re-make install and got it to run and allow users to connect. Now my problem is when more than one user connects at a time Samba seems to lock up. Checking the log files for the machines that were connected gave the following debug message: smbd/connection.c:claim_connection(208) claim_connection: tdb_store failed with error Corrupt database. do i need separate lock files for each share? can anyone shed some light on this? And I should add again that there was a problem with -lpthreads so in the make process the Makefile was edited to not use -lpthreads Dan Fenton 2011-04-28T18:56:48Z http://community.qnx.com/sf/go/post85295 Make sure io-pkt is running. Regards, -seanb On Thu, Apr 28, 2011 at 09:20:22AM -0400, Dan Fenton wrote: > Hi, > I'm trying to install Samba 6.5.8 on a QNX box running version 6.5.0 > and is's an x86 system. I've been able to get a partial build but after > doing some extra work. In the process of the install I've had to > manually install pthreads, cups, libmcrpt, and libiconv packages. Now > the samba server will run but cannot be connected to. Turning on some > debug and checking the log files I got the following messages: > > [2011/04/27 13:03:33.141910, 0] printing/pcap.c:178(pcap_cache_reload) > Unable to open printcap file /etc/printcap for read! > [2011/04/27 13:03:33.141910, 0] printing/pcap.c:178(pcap_cache_reload) > Unable to open printcap file /etc/printcap for read! > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 1 (min password > length), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 2 (password > history), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 3 (user must > logon to change password), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 4 (maximum > password age), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 5 (minimum > password age), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 6 (lockout > duration), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 7 (reset count > minutes), returning 0 > [2011/04/27 13:03:35.573538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 8 (bad lockout > attempt), returning 0 > [2011/04/27 13:03:35.574538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 9 (disconnect > time), returning 0 > [2011/04/27 13:03:35.574538, 1] > lib/account_pol.c:325(account_policy_get) > account_policy_get: tdb_fetch_uint32 failed for type 10 (refuse > machine password change), returning 0 > [2011/04/27 13:03:45.312048, 0] lib/util_sock.c:844(open_socket_in) > open_socket_in(): socket() call failed: Address family not supported > by protocol family > [2011/04/27 13:03:45.361041, 0] smbd/server.c:500(smbd_open_one_socket) > smbd_open_once_socket: open_socket_in: Address family not supported by > protocol family > [2011/04/27 13:03:45.361041, 0] lib/util_sock.c:844(open_socket_in) > open_socket_in(): socket() call failed: Address family not supported > by protocol family > [2011/04/27 13:03:45.361041, 0] smbd/server.c:500(smbd_open_one_socket) > smbd_open_once_socket: open_socket_in: Address family not supported by > protocol family > > I'm not worried about the pcap messages, the socket issues is really the > big issue. Has anyone else run into this problem and have a fix? > Also during the configuration an error involving -lpthreads was thrown. > Has this happed to anyone else and is there a fix? > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post85290 > Sean Boudreau 2011-04-28T14:11:08Z http://community.qnx.com/sf/go/post85290 Hi, I'm trying to install Samba 6.5.8 on a QNX box running version 6.5.0 and is's an x86 system. I've been able to get a partial build but after doing some extra work. In the process of the install I've had to manually install pthreads, cups, libmcrpt, and libiconv packages. Now the samba server will run but cannot be connected to. Turning on some debug and checking the log files I got the following messages: [2011/04/27 13:03:33.141910, 0] printing/pcap.c:178(pcap_cache_reload) Unable to open printcap file /etc/printcap for read! [2011/04/27 13:03:33.141910, 0] printing/pcap.c:178(pcap_cache_reload) Unable to open printcap file /etc/printcap for read! [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 1 (min password length), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 2 (password history), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 3 (user must logon to change password), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 4 (maximum password age), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 5 (minimum password age), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 6 (lockout duration), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 7 (reset count minutes), returning 0 [2011/04/27 13:03:35.573538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 8 (bad lockout attempt), returning 0 [2011/04/27 13:03:35.574538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 9 (disconnect time), returning 0 [2011/04/27 13:03:35.574538, 1] lib/account_pol.c:325(account_policy_get) account_policy_get: tdb_fetch_uint32 failed for type 10 (refuse machine password change), returning 0 [2011/04/27 13:03:45.312048, 0] lib/util_sock.c:844(open_socket_in) open_socket_in(): socket() call failed: Address family not supported by protocol family [2011/04/27 13:03:45.361041, 0] smbd/server.c:500(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address family not supported by protocol family [2011/04/27 13:03:45.361041, 0] lib/util_sock.c:844(open_socket_in) open_socket_in(): socket() call failed: Address family not supported by protocol family [2011/04/27 13:03:45.361041, 0] smbd/server.c:500(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address family not supported by protocol family I'm not worried about the pcap messages, the socket issues is really the big issue. Has anyone else run into this problem and have a fix? Also during the configuration an error involving -lpthreads was thrown. Has this happed to anyone else and is there a fix? Dan Fenton 2011-04-28T13:20:20Z http://community.qnx.com/sf/go/post85231 Im intended to do payload encryption at network layer. The encryption engine is a modified version of RC4. How to access the payload data at network layer. CRK CRK 2011-04-26T10:18:47Z http://community.qnx.com/sf/go/post84220 If you specify pci=0xX, the driver will only start on one interface. If you remove the pci=0xX from the command line, all interfaces will be started. On 11-03-23 11:46 AM, "William Luo" <community-noreply@qnx.com> wrote: > I have 3 of RTL8139C+ on one PCI bus. The PCI number are line up with 0x0, > 0x1, and 0x2. Each card can load properly with: > io-net -d rtl pci=0xX -p tcpip > > But I can't make all three cards working at the same time. The console freezes > when I load more than 1 card at the same time. Is this a driver issue? or NIC > EEPROM issue? > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post84213 > > -- Hugh Brown (613) 591-0931 ext. 2209 (voice) QNX Software Systems Ltd. (613) 591-3579 (fax) 175 Terence Matthews Cres. email: hsbrown@qnx.com Kanata, Ontario, Canada. K2M 1W8 Hugh Brown 2011-03-23T17:35:23Z http://community.qnx.com/sf/go/post84213 I have 3 of RTL8139C+ on one PCI bus. The PCI number are line up with 0x0, 0x1, and 0x2. Each card can load properly with: io-net -d rtl pci=0xX -p tcpip But I can't make all three cards working at the same time. The console freezes when I load more than 1 card at the same time. Is this a driver issue? or NIC EEPROM issue? William Luo 2011-03-23T15:46:23Z http://community.qnx.com/sf/go/post83983 Hi, I have a requirement to start "tftpd" service in mpc85xx target card. 1. I edited the "inetd.conf" and "services" files as follows and then restarted the card. /etc/inetd.conf has the configuration: tftp dgram udp nowait root /proc/boot/tftpd in.tftpd -s /software/dir /etc/services has the configuration tftp 69/udp tftpd 69/udp 2. At startup "netstat -a | grep tftp" gives: udp 0 0 *.tftp *.* 3. From the client card when I tftp to the server card and tried to get a file that is placed in the tftp root directory(the path specified in the inetd.conf) of the server, I get "Transfer timed out". 4. After this the "netstat -a | grep tftp" command in the server card does not display anything. 5. tcpdump on the server card shows, the following error: "UDP port tftp unreachable". 6. tcpdump on the client card also shows, the following error: "UDP port tftp unreachable". How can we verify that tftpd is running? Are there any other configurations to run tftpd properly? Janani K 2011-03-15T09:35:33Z http://community.qnx.com/sf/go/post83547 We have an application where multiple clients are downloading files from a QNX based file server. It is running QNX 6.3.2, using io-net. In reviewing wireshark data, it appears as though the QNX system, even though it is sending data to multiple clients, will send a burst of packets to one client, wait for an acknowledgement before sending data to any other client. We've noticed on other OS's (Windows for example) will send a burst of packets to one client, then immediately send data to the next client w/o waiting for the first to respond. If we have say a number of gigabit clients and one 100Mbit client connected to the QNX box, the 100Mbit client seems to cause overall throughput to drop for this reason. What can we do to get QNX to send data to everyone while waiting for ack's? Thanks Kevin Kevin Stallard 2011-02-26T04:17:01Z http://community.qnx.com/sf/go/post83335 Pure guess here, but maybe devn-fd can help? http://community.qnx.com/sf/discussion/do/listPosts/projects.networking/discussion.drivers.topc12934 In terms of /dev/io-net, io-net drivers are binary compatible with io-pkt. When you run them, they register as they normally would (with /dev/io-net entries). This is the driver doing the registration (not the stack). http://community.qnx.com/sf/wiki/do/viewPage/projects.networking/wiki/Drivers_wiki_page Robert craig 2011-02-18T21:34:40Z http://community.qnx.com/sf/go/post83280 > -----Message d'origine----- > De : Sean Boudreau [mailto:community-noreply@qnx.com] > Envoyé : 17 février 2011 11:20 > À : general-networking > Objet : RE: MTU Size > > > > It's 9000 not including the ethernet header + crc. Add those and you get > 9018. Does windows include those? > Are you asking me a Windows question, hahahaha. I have no idea how the Windows driver use that value. All I know is that there is only 3 settings default/4096/9016 (not sure about the 4096, don't have access to that PC at the moment). Our switch support 16K frame though, which are neither available in QNX and Windows. > -seanb > > -----Original Message----- > From: Hugh Brown [mailto:community-noreply@qnx.com] > Sent: Thu 2/17/2011 11:09 AM > To: general-networking > Subject: Re: MTU Size > > This must be something to do with the stack, as the driver can handle larger > mtu sizes. > > > On 11-02-17 9:50 AM, "Mario Charest" <community-noreply@qnx.com> > wrote: > > > Under Windows, with Intel NIC the Jumbo size that is available is 4096 > > and 9016, there isn't a way to set it to 9000, which is the maximum QNX > allows. > > > > Given that this is the exact same hardware on QNX and Windows, that > > e1000 is derived from Intel driver ( I was told ) how come the maximum > > under QNX is 9000? > > > > Or anyone know a way to get Intel driver under Windows to use jumbo > > frame of > > 9000 bytes > > > > > > > > _______________________________________________ > > > > General > > http://community.qnx.com/sf/go/post83269 > > > > > > -- > Hugh Brown (613) 591-0931 ext. 2209 (voice) > QNX Software Systems Ltd. (613) 591-3579 (fax) > 175 Terence Matthews Cres. email: hsbrown@qnx.com > Kanata, Ontario, Canada. > K2M 1W8 > > > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post83276 > > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post83278 Mario Charest 2011-02-17T16:27:10Z http://community.qnx.com/sf/go/post83278 It's 9000 not including the ethernet header + crc. Add those and you get 9018. Does windows include those? -seanb -----Original Message----- From: Hugh Brown [mailto:community-noreply@qnx.com] Sent: Thu 2/17/2011 11:09 AM To: general-networking Subject: Re: MTU Size This must be something to do with the stack, as the driver can handle larger mtu sizes. On 11-02-17 9:50 AM, "Mario Charest" <community-noreply@qnx.com> wrote: > Under Windows, with Intel NIC the Jumbo size that is available is 4096 and > 9016, there isn't a way to set it to 9000, which is the maximum QNX allows. > > Given that this is the exact same hardware on QNX and Windows, that e1000 is > derived from Intel driver ( I was told ) how come the maximum under QNX is > 9000? > > Or anyone know a way to get Intel driver under Windows to use jumbo frame of > 9000 bytes > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post83269 > > -- Hugh Brown (613) 591-0931 ext. 2209 (voice) QNX Software Systems Ltd. (613) 591-3579 (fax) 175 Terence Matthews Cres. email: hsbrown@qnx.com Kanata, Ontario, Canada. K2M 1W8 _______________________________________________ General http://community.qnx.com/sf/go/post83276 Sean Boudreau 2011-02-17T16:19:34Z http://community.qnx.com/sf/go/post83276 This must be something to do with the stack, as the driver can handle larger mtu sizes. On 11-02-17 9:50 AM, "Mario Charest" <community-noreply@qnx.com> wrote: > Under Windows, with Intel NIC the Jumbo size that is available is 4096 and > 9016, there isn't a way to set it to 9000, which is the maximum QNX allows. > > Given that this is the exact same hardware on QNX and Windows, that e1000 is > derived from Intel driver ( I was told ) how come the maximum under QNX is > 9000? > > Or anyone know a way to get Intel driver under Windows to use jumbo frame of > 9000 bytes > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post83269 > > -- Hugh Brown (613) 591-0931 ext. 2209 (voice) QNX Software Systems Ltd. (613) 591-3579 (fax) 175 Terence Matthews Cres. email: hsbrown@qnx.com Kanata, Ontario, Canada. K2M 1W8 Hugh Brown 2011-02-17T16:09:14Z http://community.qnx.com/sf/go/post83269 Under Windows, with Intel NIC the Jumbo size that is available is 4096 and 9016, there isn't a way to set it to 9000, which is the maximum QNX allows. Given that this is the exact same hardware on QNX and Windows, that e1000 is derived from Intel driver ( I was told ) how come the maximum under QNX is 9000? Or anyone know a way to get Intel driver under Windows to use jumbo frame of 9000 bytes Mario Charest 2011-02-17T14:50:24Z http://community.qnx.com/sf/go/post83259 Thanks for your suggestion. I successfully test the forwarding function in x86/QNX6.4. The traffic can travel from "en0" to serial port (ppp) and vice versa. But it seems PPP is not suitable for my case. Attached please find the topology of my system. PPP need the other peer also support ppp. But the other peer does not support ppp in my system. I am wodering for two questions: 1. Is the following possible? I write a network driver for my serial port to provide a virtual interface, say "tiger0", just like the function of "en0" or "ppp0". Then I forward traffic between "en0" and "tiger0". If this solution is OK, is it hard to write such a "pseudo" driver. The only thing it needed is to register an interface in QNX. So it can recv/send traffic to "en0". Can I refer to the Ethernet driver in offical BSP for my board? 2. io-pkt and QNX6.4. The document claims that QNX6.4.0 move to io-pkt. I start "io-pkt" in boot script. The TCP/IP runs smoothly. However, I saw "/dev/io-net" not expected "/dev/io-pkt". I also have offical QNX6.4.0 BSP for ATMEL AT91SAM926x. In the Ethernet driver, all the prefix is still "io-net-...". So I am confused. Thanks Tiger Hu Chen 2011-02-17T11:00:41Z http://community.qnx.com/sf/go/post83216 In case forwarding is not enabled on your system, check out sysctl to enable ip_forwarding flag which would allow QNX IP stack to route traffic between eth and ppp interfaces. (for that matter between any two interfaces) Regards VG On Wed, Feb 16, 2011 at 8:22 PM, Robert craig <community-noreply@qnx.com>wrote: > I believe that what you want to do is set the serial port up as a PPP link. > It will then have it's own IP address and standard IP forwarding rules > would apply. > > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post83210 > > Vineet Garg 2011-02-16T17:32:54Z http://community.qnx.com/sf/go/post83210 I believe that what you want to do is set the serial port up as a PPP link. It will then have it's own IP address and standard IP forwarding rules would apply. Robert craig 2011-02-16T14:52:08Z http://community.qnx.com/sf/go/post83193 All, I have an embedded device which has an Ethernet interface and serial port. I want this device to work like a "router", that is, 1.The device get IP packet from the Ethernet, it the dest IP is not the device itself. It just foward the packet to the serial port. 2. The device get data from the serial port (supposed to be IP packet) and send this data (IP packet) to Ethernet. 3. The device has its own IP and can response ping and ARP from the Ethernet interface. Now I have QNX running, Ethernet and Serial port are all OK. So how can I realize the forwarding function? BPF or PF. I just read the doc for "packet filtering" and has no idea so far. I will not handle the Ethernet header. So BPF is not necessary? Thanks in advance for any clues. Hu Chen 2011-02-16T03:44:12Z http://community.qnx.com/sf/go/post82889 Thanks for the reply but then I need to know exactly when TCP stacks seeds it? Is there any other way we can get directly ip conflict issue from TCP nishant mishra 2011-02-04T06:54:31Z http://community.qnx.com/sf/go/post82862 The tcpip stack seeds the 'duplicate IP ..' message. You can get the count from 'netstat -parp | grep duplicates' but you don't get the addresses. Not sure about qnet. -seanb -----Original Message----- From: nishant mishra [mailto:community-noreply@qnx.com] Sent: Thu 2/3/2011 4:23 AM To: general-networking Subject: How to tap /dev/slog I need to detect ip conflict on my QNX based networking system I tried reading /dev/slog and got "duplicate IP address 128.0.0.3 sent from link address 00:21:70:d2:af:34" into slog. 1) Need to know how this gets filled and by whom? 2) Is there any way to directly get hold of ip conflict Also "npm-qnet(L4): lr_inbound(): rxd pkt type 2 with duplicate name/domain, staying online". gets in slog when there is any other conflict on the network _______________________________________________ General http://community.qnx.com/sf/go/post82859 Sean Boudreau 2011-02-03T14:25:51Z http://community.qnx.com/sf/go/post82859 I need to detect ip conflict on my QNX based networking system I tried reading /dev/slog and got “duplicate IP address 128.0.0.3 sent from link address 00:21:70:d2:af:34” into slog. 1) Need to know how this gets filled and by whom? 2) Is there any way to directly get hold of ip conflict Also “npm-qnet(L4): lr_inbound(): rxd pkt type 2 with duplicate name/domain, staying online”. gets in slog when there is any other conflict on the network nishant mishra 2011-02-03T09:23:08Z http://community.qnx.com/sf/go/post82764 Hello, I need to write a wifi driver for a SPI connected module. So I need to access networking source repository to start with template driver source code. If I have understood well, I have to register to QNX source project to have access to the source repository. That's what I did about 2 weeks ago. My request is still pending. What's wrong ? Best regards, Nicolas Nicolas Pinault 2011-01-31T09:54:26Z http://community.qnx.com/sf/go/post82706 > ooops... sleepy eyes made me type DGRAM... and IPPROTO_ICMP is defined as 1 > anyway... > Thanks for the utils/p/ping tip.. will try... How can I get the source code of such utility? Can you give me the link? Hu Chen 2011-01-28T07:17:02Z http://community.qnx.com/sf/go/post82632 It does work . Thank you Subra arya 2011-01-25T16:19:26Z http://community.qnx.com/sf/go/post82628 When copying .boot files like that remember to run sync before you reboot, especially on the qnx6 fs On 11-01-25 10:36 AM, Mario Charest wrote: > >> -----Message d'origine----- >> De : Subra arya [mailto:community-noreply@qnx.com] >> Envoyé : 25 janvier 2011 08:40 >> À : general-networking >> Objet : Re: RE: QNX 6.4.1 dual core >> >> I did try that too ie copying /boot/fs/qnxbasesmp.if /.boot . > I assume you rebooted after you did that? Maybe the second core is disable in the BIOS? > >> My pidin in out put remains same . >> I know i am running a dual core . >> So just by looking at pidin in output i am not able to say if my os is taking >> advantage of all the cpu cores or not . >> Any other info is helpful . >> >> thanks >> subra >> >> >> >> _______________________________________________ >> >> General >> http://community.qnx.com/sf/go/post82615 >> > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post82627 -- cburgess@qnx.com Colin Burgess 2011-01-25T15:41:10Z http://community.qnx.com/sf/go/post82627 > -----Message d'origine----- > De : Subra arya [mailto:community-noreply@qnx.com] > Envoyé : 25 janvier 2011 08:40 > À : general-networking > Objet : Re: RE: QNX 6.4.1 dual core > > I did try that too ie copying /boot/fs/qnxbasesmp.if /.boot . I assume you rebooted after you did that? Maybe the second core is disable in the BIOS? > My pidin in out put remains same . > I know i am running a dual core . > So just by looking at pidin in output i am not able to say if my os is taking > advantage of all the cpu cores or not . > Any other info is helpful . > > thanks > subra > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post82615 > Mario Charest 2011-01-25T15:36:47Z http://community.qnx.com/sf/go/post82615 I did try that too ie copying /boot/fs/qnxbasesmp.if /.boot . My pidin in out put remains same . I know i am running a dual core . So just by looking at pidin in output i am not able to say if my os is taking advantage of all the cpu cores or not . Any other info is helpful . thanks subra Subra arya 2011-01-25T13:40:11Z http://community.qnx.com/sf/go/post82596 By default the installation process install the single core version. To install the smp version cp /boot/fs/qnxsmpbase.ifs /.boot > -----Message d'origine----- > De : Subra arya [mailto:community-noreply@qnx.com] > Envoyé : 24 janvier 2011 16:26 > À : general-networking > Objet : QNX 6.4.1 dual core > > Hello All , > > When we install the qnx 6.4.1 os on an intel dual core processor , would it by > default use only one cpu or would it configure to take advantage of both > CPU's . I have done just a regular install with type > 79 file system. > > The pidin in listing relevant o/p says > > Processor1: 66222 Pentium III Stepping 10 2825MHz FPU > > does that mean it is aware of only one cpu > > thanks > subra > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post82595 > Mario Charest 2011-01-24T21:53:24Z http://community.qnx.com/sf/go/post82595 Hello All , When we install the qnx 6.4.1 os on an intel dual core processor , would it by default use only one cpu or would it configure to take advantage of both CPU's . I have done just a regular install with type 79 file system. The pidin in listing relevant o/p says Processor1: 66222 Pentium III Stepping 10 2825MHz FPU does that mean it is aware of only one cpu thanks subra Subra arya 2011-01-24T21:25:54Z http://community.qnx.com/sf/go/post82594 The pidin in listing relevant o/p says Processor1: 66222 Pentium III Stepping 10 2825MHz FPU does that mean it is aware of only one cpu thanks subra Subra arya 2011-01-24T20:43:20Z http://community.qnx.com/sf/go/post82590 The pidin in listing relevant o/p says Processor1: 66222 Pentium III Stepping 10 2825MHz FPU does that mean it is aware of only one cpu thanks subra Subra arya 2011-01-24T19:51:14Z http://community.qnx.com/sf/go/post82589 Look at the output of 'pidin in' to see how many cores the os is aware of. -seanb -----Original Message----- From: Subra arya [mailto:community-noreply@qnx.com] Sent: Mon 1/24/2011 2:22 PM To: general-networking Subject: QNX 6.4.1 on dual core Hello All , When we install the qnx 6.4.1 os on an intel dual core processor , would it by default use only one cpu or would it configure to take advantage of both CPU's . I have done just a regular install with type 79 file system. Please let me know Thanks Subra _______________________________________________ General http://community.qnx.com/sf/go/post82587 Sean Boudreau 2011-01-24T19:31:34Z http://community.qnx.com/sf/go/post82588 What does 'pidin in' say? On 11-01-24 02:22 PM, Subra arya wrote: > Hello All , > > When we install the qnx 6.4.1 os on an intel dual core processor , > would it by default use only one cpu or would it configure to take > advantage of both CPU's . I have done just a regular install with type > 79 file system. > Please let me know > Thanks > Subra > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post82587 > Patrik Lahti 2011-01-24T19:31:05Z http://community.qnx.com/sf/go/post82587 Hello All , When we install the qnx 6.4.1 os on an intel dual core processor , would it by default use only one cpu or would it configure to take advantage of both CPU's . I have done just a regular install with type 79 file system. Please let me know Thanks Subra Subra arya 2011-01-24T19:22:10Z http://community.qnx.com/sf/go/post82524 Hello. I'm looking for a USB miniCard that would be used on our HW under QNX 6.4.1. I've found the "QCOM Q802XKG" using a Ralink Ralink RT2571W MAC/BBP with RT2528 Transceiver chipset. Attached are the card's specifications I would know if this minicard is supported by one of the Ralink USB drivers of QNX, devnp-rum or devnp-ural ? Thanks you very much. Guillaume. Guillaume Varlet 2011-01-21T10:24:14Z http://community.qnx.com/sf/go/post81149 After further research I have found an online book "The Boost C++ Libraries" by Boris Schaling translated by Andreas Masur published on en.highscore.de that has a good section on extending the asio library http://en.highscore.de/cpp/boost/asio.html#asio_extensions Using the information presented in the book and asio library documentation, it would be feasible to implement asio service and io_objects to integrate zmq into the asio framework. ref: asio.sourceforge.net and boost.org Mike Mike Robins 2011-01-05T04:09:18Z http://community.qnx.com/sf/go/post81056 That's an io-net driver, so you're running the shim under io-pkt. After putting the laptop to sleep, then waking it up again, if you umount the driver, then mount it again, does it start to work again? For your second question, you need to figure out if you have a driver problem or a protocol problem (shot in the dark - probably driver problem). You can diagnose driver problems with the "nicinfo" and "ifconfig -v" commands. You can also use the "verbose=X" options to the driver, and hopefully they might log more information (eg link up down) to the sloginfo output. What you're looking for - does it report a good link? Are the rx pkt counts going up? Andrew Boyd 2011-01-04T16:33:09Z http://community.qnx.com/sf/go/post80779 We are current running QNX 6.4.1 with the ns83815 drivers Any suggestions on how to attack this problem would be greatly appreciated. We have multiple ways of causing the TCPIP subsystem to fail, these are but 2 examples: 1) If we directly connect a laptop to the Ethernet card on our pc104 stack (with the appropriate crossover) and establish a connection to our data acquisition application and then put the laptop to "sleep" mode, we seem to completely lose **ALL** TCP/IP functionality on the QNX box -- which will only be restored with a complete reboot. 2) connect QNX to operate with a wireless router on the lan side -- log a laptop into the router and then setup an application to talk with QNX (via our DAS app, telnet, ftp, etc) Move the laptop to a fringe location and wait for the wireless session to break. Once the wireless session breaks with QNX, the TCPIP subsystem then random breaks approximently 20% of the time in the QNX box -- again requiring a complete reboot. Is there any way to detect if the TCP drivers have failed and perhaps reboot these alone? We are completely stumped by this. Thanks James Kelley James Kelley 2010-12-31T18:14:25Z http://community.qnx.com/sf/go/post80208 Hello, I have communications problems in a link between two hosts in different tcp/ip networks (i'm using a router to do this communication). My Server is the IP 192.168.11.1 (QNX host) and the client host is 192.168.0.1 (non qnx host - RTU) The router have the IP address 192.168.11.252, so I added the route #route add -net 192.168.0.0 -netmask 255.255.255.248 192.168.11.252 First, I assumed a fail in the fiber link between the router and the RTU. So I proved with a "all-day" ping to get some information about the problem. The result was, several lost of pings, whith the "no route to host" message. Then, I've created a virtual interface in the router (An allways alive interface or network from the point of view of the other networks). The result of the ping test, was the same for my virtual interface. Therefore, I can't assume a problem in the fiber link. And, as the error message says, the problem is a lost route for the network 192.168.0.0/29. The same test, pinging to the router from my host, was always successfuly. Is possible, that my QNX server losses routes for periods of about 40 seconds o less? Internet: Destination Gateway Flags Refs Use Mtu Interface 10.210.40/22 link#5 UC 49 0 - en3 10.210.40.10 00:a0:f4:01:00:40 UHLc 1 55732081 - en3 10.210.40.11 00:a0:f4:00:0a:1f UHLc 1 5224829 - en3 10.210.40.12 00:a0:f4:01:00:44 UHLc 1 55693107 - en3 10.210.40.14 00:a0:f4:01:00:41 UHLc 1 55730625 - en3 10.210.40.15 00:a0:f4:01:00:46 UHLc 1 55697045 - en3 10.210.40.16 00:a0:f4:01:44:96 UHLc 1 55705008 - en3 10.210.40.21 00:a0:f4:01:00:48 UHLc 1 55724202 - en3 10.210.40.23 00:a0:f4:01:00:43 UHLc 1 5236297 - en3 10.210.40.36 00:a0:f4:04:00:a4 UHLc 1 16804378 - en3 10.210.40.70 00:a0:f4:01:41:f6 UHLc 1 27221310 - en3 10.210.40.71 00:a0:f4:01:56:7b UHLc 1 27213964 - en3 10.210.40.74 00:a0:f4:01:41:f5 UHLc 1 27287201 - en3 10.210.40.75 00:a0:f4:01:56:fb UHLc 1 27224002 - en3 10.210.40.82 00:a0:f4:01:56:53 UHLc 1 27300330 - en3 10.210.40.83 00:a0:f4:01:56:fc UHLc 1 27280735 - en3 10.210.40.108 00:a0:f4:01:37:2d UHLc 1 17493316 - en3 10.210.40.118 00:a0:f4:41:6f:2e UHLc 1 33207717 - en3 10.210.40.151 00:a0:f4:01:16:74 UHLc 1 24887446 - en3 10.210.40.254 00:0e:0c:7f:c0:b5 UHLc 0 2 - en3 10.210.41.7 00:a0:f4:01:a5:f9 UHLc 1 8823528 - en3 10.210.41.8 00:a0:f4:01:91:b5 UHLc 1 5564906 - en3 10.210.41.15 00:a0:f4:01:7b:5f UHLc 1 8543950 - en3 10.210.41.16 00:a0:f4:01:7b:33 UHLc 1 5183769 - en3 10.210.41.18 00:a0:f4:04:1f:07 UHLc 1 8545354 - en3 10.210.41.19 00:a0:f4:01:a5:f7 UHLc 1 8309208 - en3 10.210.41.20 00:a0:f4:01:90:85 UHLc 1 5203971 - en3 10.210.41.22 00:a0:f4:00:15:ac UHLc 1 2405227 - en3 10.210.41.25 00:a0:f4:01:91:b7 UHLc 1 4914377 - en3 10.210.41.26 00:a0:f4:01:90:8a UHLc 1 6445966 - en3 10.210.41.27 00:a0:f4:01:90:88 UHLc 1 6446406 - en3 10.210.41.28 00:a0:f4:01:90:62 UHLc 1 209450 - en3 10.210.41.29 00:a0:f4:01:90:63 UHLc 1 210362 - en3 10.210.41.246 00:00:54:14:70:ab UHLc 1 131239584 - en3 10.210.42.5 00:a0:f4:01:97:69 UHLc 1 37821135 - en3 10.210.42.6 00:a0:f4:01:97:53 UHLc 1 41906247 - en3 10.210.42.7 00:a0:f4:01:97:56 UHLc 1 41673128 - en3 10.210.42.8 00:a0:f4:01:97:54 UHLc 1 37313766 - en3 10.210.42.9 00:a0:f4:01:97:55 UHLc 1 5024141 - en3 10.210.42.22 00:a0:f4:01:97:5b UHLc 1 1309884 - en3 10.210.42.23 00:a0:f4:01:97:5a UHLc 1 38548278 - en3 10.210.42.34 00:a0:f4:01:97:5e UHLc 1 37388793 - en3 10.210.42.42 00:a0:f4:01:a0:1c UHLc 1 4903365 - en3 10.210.42.43 00:a0:f4:01:a0:37 UHLc 1 24174293 - en3 10.210.42.44 00:a0:f4:01:a0:33 UHLc 1 24174516 - en3 10.210.42.45 00:a0:f4:01:a0:1e UHLc 1 23463998 - en3 10.210.42.46 00:a0:f4:01:a0:24 UHLc 1 24173733 - en3 10.210.42.47 00:a0:f4:01:9f:f6 UHLc 1 5024031 - en3 10.210.42.244 00:1b:21:23:0f:4a UHLc 0 4 - lo0 10.210.42.246 00:40:f4:89:c2:dd UHLc 0 2 - en3 localhost.localdom localhost.localdom UH 0 3971 33212 lo0 192.168/29 192.168.11.252 UGS 6 22657911 - en1 192.168.0.8/29 192.168.11.252 UGS 0 0 - en1 192.168.10 link#2 UC 5 0 - en0 EL-01 00:0b:ab:1a:c0:e5 UHLc 0 13904 - lo0 EL-02 00:0b:ab:1a:c0:fa UHLc 1 544324 - en0 EL-05 00:0b:ab:1a:c7:b9 UHLc 2 6975 - en0 NASA link#2 UHLc 1 29663 - en0 192.168.10.255 link#2 UHLc 2 2603320138 - en0 192.168.11 link#3 UC 2 0 - en1 EL-01_2 00:0b:ab:1a:c0:e6 UHLc 0 4 - lo0 192.168.11.252 00:1f:9e:39:ec:b5 UHLc 3 95529 - en1 192.168.12 link#4 UC 1 0 - en2 192.168.12.255 link#4 UHLc 1 110664 - en2 192.168.254 192.168.11.252 UGS 0 86592 - en1 Internet6: Destination Gateway Flags Refs Use Mtu Interface localhost.localdom localhost.localdom UH 0 0 33212 lo0 fe80:1:: fe80:1::1 U 0 0 33212 lo0 ff01:: localhost.localdom U 0 0 33212 lo0 ff02:1:: fe80:1::1 UC 0 0 33212 lo0 Thanks in advance Gabriel Gabriel Cane 2010-12-23T14:59:15Z http://community.qnx.com/sf/go/post79725 Mike, Mike Robins wrote: > Armin, > > ZeroMQ does look good and I have looked into using it before, but our > app is rather tied to Boost::asio. We make extensive use of the > asynchronous call back mechanisms. > > Have you seen any adaptations of asio to use ZeroMQ as the underlying > comms? Both systems are written in C++ so it could be possible that someone has done that. But you should post this question in the 0MQ mailing list ... --Armin >> the middleware ZeroMQ ( http://www.zeromq.org ) could be an alternative >> to fiddeling around with io-pkt ... > -- Mike > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post79701 > > Armin Steinhoff 2010-12-22T11:32:00Z http://community.qnx.com/sf/go/post79701 Armin, ZeroMQ does look good and I have looked into using it before, but our app is rather tied to Boost::asio. We make extensive use of the asynchronous call back mechanisms. Have you seen any adaptations of asio to use ZeroMQ as the underlying comms? > the middleware ZeroMQ ( http://www.zeromq.org ) could be an alternative > to fiddeling around with io-pkt ... -- Mike Mike Robins 2010-12-22T00:46:13Z http://community.qnx.com/sf/go/post79392 Are there any QNX preferences to TCP server architecture? I'm talking about a number of (remote) clients using TCP to make requests to a TCP server application that is listening on one TCP port, hosted on QNX. There might be up to a couple of hundred TCP sockets active at a time (though it could be restricted to less). The kind of preference I mean is, eg, is it better to have one thread handle each client connection or would it be better to use, say switch() to use one thread to listen for activity on all sessions? Or some combination of the two. Obviously there are costs/limits associated with each approach and there is a potential performance is with using switch(). Is there an underlying message-based API that would be suitable at the application level? Chris Trobridge 2010-12-20T16:06:37Z http://community.qnx.com/sf/go/post79126 Mike, the middleware ZeroMQ ( http://www.zeromq.org ) could be an alternative to fiddeling around with io-pkt ... --Armin Mike Robins wrote: > I am porting an application to QNX (I'm relatively new to QNX) that utilizes > TCP/IP sockets to implement client-server IPC. I want to maintain the ability > to use TCP/IP because my test server only runs on Windows. > > I would like my application to utilise native communications when client and > server are both running on QNX. My reading seems to indicate that io-pkt can > utilise TCP or QNET so I would like to have my application talk to the io-pkt > and open a connection across either protocol. I'll know which protocol would > be needed at run time so can give suitably different addresses or what ever. > > I want the application to treat the connection as a socket regardless of > the transport mechanism. That is, I want a socket interface implemented > across a channel rather than IP. Isn't this what io-pkt does? > > I don't want to have to re-implement the server as a resource manager. > > I can't find straight forward examples of using io-pkt. > > > Have I got the wrong end of the stick? Is the situation that io-pkt either > resource manages the /dev/socket OR /dev/net namespace and the two are accessed > differently: TCP/IP through libsocket and QNET through libc? > > In which case is io-pkt smart enough to notice that I am opening a local > TCP socket and implement it really efficiently? > > Any help would be appreciated. > > PS my application uses the Boost::asio library > > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post79119 > > Armin Steinhoff 2010-12-17T10:48:11Z http://community.qnx.com/sf/go/post79119 I am porting an application to QNX (I'm relatively new to QNX) that utilizes TCP/IP sockets to implement client-server IPC. I want to maintain the ability to use TCP/IP because my test server only runs on Windows. I would like my application to utilise native communications when client and server are both running on QNX. My reading seems to indicate that io-pkt can utilise TCP or QNET so I would like to have my application talk to the io-pkt and open a connection across either protocol. I'll know which protocol would be needed at run time so can give suitably different addresses or what ever. I want the application to treat the connection as a socket regardless of the transport mechanism. That is, I want a socket interface implemented across a channel rather than IP. Isn't this what io-pkt does? I don't want to have to re-implement the server as a resource manager. I can't find straight forward examples of using io-pkt. Have I got the wrong end of the stick? Is the situation that io-pkt either resource manages the /dev/socket OR /dev/net namespace and the two are accessed differently: TCP/IP through libsocket and QNET through libc? In which case is io-pkt smart enough to notice that I am opening a local TCP socket and implement it really efficiently? Any help would be appreciated. PS my application uses the Boost::asio library Mike Robins 2010-12-17T07:29:26Z http://community.qnx.com/sf/go/post78666 Can one of the admins please respond why I get this errors? I have tried resetting passwords etc. Has not helped. > svn checkout --username subodh.harmalkar@jhuapl.edu http://community.qnx.com/svn/repos/core_networking/tags/6.4.0/GA Authentication realm: <http://community.qnx.com:80> Authorization Realm Password for 'subodh.harmalkar@jhuapl.edu': Authentication realm: <http://community.qnx.com:80> Authorization Realm Username: Subodh Harmalkar 2010-12-14T18:29:50Z http://community.qnx.com/sf/go/post78450 Hi, You're talking about RFC 6106, which is currently not supported. You can use DHCPv6 for getting DNSv6 servers or use the DNSv4 servers to look up AAAA records (in a dual stack network). The latter is probably the more common deployment scenario anyways. DHCPv6 support is implemented but not yet released. You can perhaps request it through the appropriate support channels. Cheers! /P PS Strange, I didn't see this posting before, apologies for the late answer. On 10-12-13 11:34 AM, Owen James wrote: > Any ideas? > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post78439 > Patrik Lahti 2010-12-13T17:06:34Z http://community.qnx.com/sf/go/post78439 Any ideas? Owen James 2010-12-13T16:34:20Z http://community.qnx.com/sf/go/post77789 Thanks Glenn, I didn't realize you had already answered. PS. snap length zero can also be used. On 10-12-08 08:01 AM, Glenn Schmottlach wrote: > Wireshark doesn't run on QNX. What you do is use tcpdump (which runs > under QNX) to capture a trace and write it to a file that ends in > "*.pcap". Transfer that file to your PC and open it with Wireshark. > > I think the command I usually use is something like: > > tcpdump -i en0 -s 65535 -w trace.pcap > > Hope that helps. > > Glenn > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post77760 > Patrik Lahti 2010-12-08T14:55:31Z http://community.qnx.com/sf/go/post77787 Hi, There's no Wireshark port for QNX AFAIK, but you can use tcpdump to grab a pcap file on target. Then you can use Wireshark on another OS to read the capture. Hope this helps! /P On 10-12-08 04:40 AM, swati jaiswal wrote: > Hi > > I need wireshark software for QNX 6.4, so can you please tell from where > I can get it and How to install it on QNX 6.4 > > > Thanks > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post77752 > Patrik Lahti 2010-12-08T14:54:41Z http://community.qnx.com/sf/go/post77760 Wireshark doesn't run on QNX. What you do is use tcpdump (which runs under QNX) to capture a trace and write it to a file that ends in "*.pcap". Transfer that file to your PC and open it with Wireshark. I think the command I usually use is something like: tcpdump -i en0 -s 65535 -w trace.pcap Hope that helps. Glenn Glenn Schmottlach 2010-12-08T13:01:01Z http://community.qnx.com/sf/go/post77752 Hi I need wireshark software for QNX 6.4, so can you please tell from where I can get it and How to install it on QNX 6.4 Thanks swati jaiswal 2010-12-08T09:40:23Z http://community.qnx.com/sf/go/post77646 On Tue, Dec 07, 2010 at 08:33:08AM -0500, Matt Blackburn wrote: > I have a ppp configured connection and ifconfig reports the following: > > ppp0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1600 > inet 166.217.73.119 -> 10.0.0.1 netmask 0xffff0000 > > Can someone help me understand the 166.217.73.119 -> 10.0.0.1 please. Those are the addresses of either end of the ppp link (local -> remote). -seanb Sean Boudreau 2010-12-07T16:50:49Z http://community.qnx.com/sf/go/post77584 I have a ppp configured connection and ifconfig reports the following: ppp0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1600 inet 166.217.73.119 -> 10.0.0.1 netmask 0xffff0000 Can someone help me understand the 166.217.73.119 -> 10.0.0.1 please. Matt Blackburn 2010-12-07T13:33:07Z http://community.qnx.com/sf/go/post76275 Hi, A customer is asking if we put our io-net or io-pkt IPV4 stacks through any form of external compliance testing. Thanks, Garry Garry Bleasdale 2010-11-29T15:46:24Z http://community.qnx.com/sf/go/post76166 Thanks very much for the detailed response Dave. Regards, Garry Garry Bleasdale Field Application Engineer QNX Software Systems (UK + Ireland) Email - gbleasdale@qnx.com Tel - +44 (0) 1223 862077 www.qnx.com -----Original Message----- From: Dave Brown [mailto:community-noreply@qnx.com] Sent: 25 November 2010 22:11 To: general-networking Subject: RE: SNMP Question When they are saying they don't need to recompile their code for each product are they referring to dynamically extending the agent? The base SNMP agent which is supplied with 6.5 is not dynamically extensible and does not have a toolkit which can generate skeleton code for extending the MIB support of the agent. The only way to extend this agent is to modify the agent code. There is a SNMP agent in the foundry "net-snmp" which I believe does have some of this capability. I think it offers both static and dynamic extensibility via a toolkit, but I have not used it, so I am can't say from experience how extensive the support is. In this case the toolkit would generate skeleton code which can be filled in with the code to set or get the MIB database entry data you are interested in. There are also 3rd parties such as NuDesign and SNMP Research which offer much more feature rich packages. Thanks Dave > -----Original Message----- > From: Garry Bleasdale [mailto:community-noreply@qnx.com] > Sent: November 25, 2010 5:22 AM > To: general-networking > Subject: SNMP Question > > > Hi, > > A customer has a question regarding our SNMP: > > Our products offer SNMP Agents. We generate our SNMP tables from an XML > file so the SNMP code does not need to be recompiled for each product. > The Trek SNMP library supplies us with basic MIB2 support and allows us > to hook in a subtree for our own use (in the enterprise specific are of > OID space). > > Is this kind of thing possible with QNX's SNMP Agent? > > Thanks, > > Garry > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post75943 _______________________________________________ General http://community.qnx.com/sf/go/post76070 Garry Bleasdale 2010-11-26T17:34:28Z http://community.qnx.com/sf/go/post76070 When they are saying they don't need to recompile their code for each product are they referring to dynamically extending the agent? The base SNMP agent which is supplied with 6.5 is not dynamically extensible and does not have a toolkit which can generate skeleton code for extending the MIB support of the agent. The only way to extend this agent is to modify the agent code. There is a SNMP agent in the foundry "net-snmp" which I believe does have some of this capability. I think it offers both static and dynamic extensibility via a toolkit, but I have not used it, so I am can't say from experience how extensive the support is. In this case the toolkit would generate skeleton code which can be filled in with the code to set or get the MIB database entry data you are interested in. There are also 3rd parties such as NuDesign and SNMP Research which offer much more feature rich packages. Thanks Dave > -----Original Message----- > From: Garry Bleasdale [mailto:community-noreply@qnx.com] > Sent: November 25, 2010 5:22 AM > To: general-networking > Subject: SNMP Question > > > Hi, > > A customer has a question regarding our SNMP: > > Our products offer SNMP Agents. We generate our SNMP tables from an XML > file so the SNMP code does not need to be recompiled for each product. > The Trek SNMP library supplies us with basic MIB2 support and allows us > to hook in a subtree for our own use (in the enterprise specific are of > OID space). > > Is this kind of thing possible with QNX's SNMP Agent? > > Thanks, > > Garry > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post75943 Dave Brown 2010-11-25T22:11:14Z http://community.qnx.com/sf/go/post76048 Cheers Patrick. Regards, Garry Garry Bleasdale Field Application Engineer QNX Software Systems (UK + Ireland) Email - gbleasdale@qnx.com Tel - +44 (0) 1223 862077 www.qnx.com -----Original Message----- From: Patrik Lahti [mailto:community-noreply@qnx.com] Sent: 25 November 2010 14:32 To: general-networking Subject: Re: SNMP Question Hey Garry, It's software, so anything is possible :-) Unfortunately, without spending some time, I don't know off hand what it would take to do what's described. Cheers! /P On 10-11-25 05:22 AM, Garry Bleasdale wrote: > > Hi, > > A customer has a question regarding our SNMP: > > Our products offer SNMP Agents. We generate our SNMP tables from an XML > file so the SNMP code does not need to be recompiled for each product. > The Trek SNMP library supplies us with basic MIB2 support and allows us > to hook in a subtree for our own use (in the enterprise specific are of > OID space). > > Is this kind of thing possible with QNX's SNMP Agent? > > Thanks, > > Garry > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post75943 > _______________________________________________ General http://community.qnx.com/sf/go/post75961 Garry Bleasdale 2010-11-25T21:01:30Z http://community.qnx.com/sf/go/post75961 Hey Garry, It's software, so anything is possible :-) Unfortunately, without spending some time, I don't know off hand what it would take to do what's described. Cheers! /P On 10-11-25 05:22 AM, Garry Bleasdale wrote: > > Hi, > > A customer has a question regarding our SNMP: > > Our products offer SNMP Agents. We generate our SNMP tables from an XML > file so the SNMP code does not need to be recompiled for each product. > The Trek SNMP library supplies us with basic MIB2 support and allows us > to hook in a subtree for our own use (in the enterprise specific are of > OID space). > > Is this kind of thing possible with QNX's SNMP Agent? > > Thanks, > > Garry > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post75943 > Patrik Lahti 2010-11-25T14:31:54Z http://community.qnx.com/sf/go/post75943 Hi, A customer has a question regarding our SNMP: Our products offer SNMP Agents. We generate our SNMP tables from an XML file so the SNMP code does not need to be recompiled for each product. The Trek SNMP library supplies us with basic MIB2 support and allows us to hook in a subtree for our own use (in the enterprise specific are of OID space). Is this kind of thing possible with QNX's SNMP Agent? Thanks, Garry Garry Bleasdale 2010-11-25T10:22:02Z http://community.qnx.com/sf/go/post75353 Does rtsold solicit IPv6 routers for DNSv6 information? If so, where does it store these values? Owen James 2010-11-19T18:46:48Z http://community.qnx.com/sf/go/post74751 Problem solved by /lib/libsnmp.so from QNX6.5.0 Are there some source examples how to use libsnmp functions? Thank you. Leonid Khait 2010-11-16T07:18:37Z http://community.qnx.com/sf/go/post74630 Hi Robert, Good to hear from you and thanks very much for the detailed response. Regards, Garry Garry Bleasdale Field Application Engineer QNX Software Systems (UK + Ireland) Email - gbleasdale@qnx.com Tel - +44 (0) 1223 862077 www.qnx.com -----Original Message----- From: Robert craig [mailto:community-noreply@qnx.com] Sent: 12 November 2010 19:20 To: general-networking Subject: Re: RE: Qnet across SPI? A pseudo-Ethernet driver approach would be the most straight forward thing to do. Making it look like an Ethernet driver has the advantage of letting you use IP or QNET or any other protocol for the upper layers seamlessly. The serial RapidIO package done a few years back used this method (albeit with io-net and not io-pkt). You're not going to get the best performance since you'll be pulling in QNET's reliability layer when you don't need it, but bypassing the layer is not for the faint of heart. In terms of overhead, you could just send Ethernet packets over the interface as is or, if there's too much overhead, simply remove the Ethernet header before sending the packet and add it back after receiving a packet (before sending it up the stack). The MAC that you select for each host could uniquely determine the appropriate SPI interface to transmit on. Robert. _______________________________________________ General http://community.qnx.com/sf/go/post74512 Garry Bleasdale 2010-11-15T15:20:28Z http://community.qnx.com/sf/go/post74606 I need ed to implement SNMP traps.Traps is to be send from MPC 8548cds board to NMS running on some other PC . I have with me my custom MIB module. How to load/integrate this MIB module in QNX 6. 4 os running on the MPC 8548CDS board?? Till now I have done following things:- 1)running a trap reciver on some PC. 2) I used snmptrap command fom QNX 6.4 to send a trap. . I am able to receive the trap on the other side. MIB module used to send this trap is in perhaps /etc/mib.txt. Problem I am facing is ,custom MIB module is not getting loaded with the QNX os. Plz let me know the procedure to do this.Also how I can start QNX agent?? Regards, Ravish Kumar Ravish Kumar 2010-11-15T14:07:35Z http://community.qnx.com/sf/go/post74598 I try to use snmpget, snmptest commands in QNX6.4.1 : #snmpget -v 1 host_name public system.SysUpTime.0 and received info, but with Memory fault (core dumped) in final. snmpget.coreinfo in Attachmetn Please describe the reason of the problem. Are there some source examples how to use libsnmp functions? Thanks. Leonid Khait 2010-11-15T12:40:39Z http://community.qnx.com/sf/go/post74512 A pseudo-Ethernet driver approach would be the most straight forward thing to do. Making it look like an Ethernet driver has the advantage of letting you use IP or QNET or any other protocol for the upper layers seamlessly. The serial RapidIO package done a few years back used this method (albeit with io-net and not io-pkt). You're not going to get the best performance since you'll be pulling in QNET's reliability layer when you don't need it, but bypassing the layer is not for the faint of heart. In terms of overhead, you could just send Ethernet packets over the interface as is or, if there's too much overhead, simply remove the Ethernet header before sending the packet and add it back after receiving a packet (before sending it up the stack). The MAC that you select for each host could uniquely determine the appropriate SPI interface to transmit on. Robert. Robert craig 2010-11-12T19:19:58Z http://community.qnx.com/sf/go/post74431 Cheers Patrik. Regards, Garry Garry Bleasdale Field Application Engineer QNX Software Systems (UK + Ireland) Email - gbleasdale@qnx.com Tel - +44 (0) 1223 862077 www.qnx.com -----Original Message----- From: Patrik Lahti [mailto:community-noreply@qnx.com] Sent: 10 November 2010 14:47 To: general-networking Subject: Re: Qnet across SPI? SPI is just a bus. But if you can run Ethernet on top of it I assume you can run Qnet over it. Or you may be able to enhance Qnet to work over another transport. /P On 10-11-10 06:31 AM, Garry Bleasdale wrote: > Hi, > > A customer plans to connect three ARM9 based CPUs using SPI and is > asking if it is possible to use Qnet? > > Thanks, > > Garry? > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post74180 > _______________________________________________ General http://community.qnx.com/sf/go/post74198 Garry Bleasdale 2010-11-12T09:51:02Z http://community.qnx.com/sf/go/post74198 SPI is just a bus. But if you can run Ethernet on top of it I assume you can run Qnet over it. Or you may be able to enhance Qnet to work over another transport. /P On 10-11-10 06:31 AM, Garry Bleasdale wrote: > Hi, > > A customer plans to connect three ARM9 based CPUs using SPI and is > asking if it is possible to use Qnet? > > Thanks, > > Garry? > > > > _______________________________________________ > > General > http://community.qnx.com/sf/go/post74180 > Patrik Lahti 2010-11-10T14:46:54Z http://community.qnx.com/sf/go/post74180 Hi, A customer plans to connect three ARM9 based CPUs using SPI and is asking if it is possible to use Qnet? Thanks, Garry? Garry Bleasdale 2010-11-10T11:31:05Z http://community.qnx.com/sf/go/post73982 I am *trying* to come up with firewall rules to allow DHCP to work reliably from my firewalled target system. The simplest firewall (when enabled wth 'pfctl -e') works perfectly with dhcp.client: ## Scrub scrub in all ## Filter Rules pass in log all keep state pass out log all keep state Monitoring /dev/pflog0 I see the following when I make a dhcp.client request: > dhcp.client -um rule 1/0(match): pass out on en0: 0.0.0.0.68 > 255.255.255.255.67: BOOTP/DHCP, Request [|bootp] rule 0/0(match): pass in on en0: 10.30.174.2.67 > 10.30.175.133.68: BOOTP/DHCP, Reply, length 363 rule 0/0(match): pass in on en0: 10.30.174.3.67 > 10.30.175.133.68: BOOTP/DHCP, Reply, length 363 rule 0/0(match): pass in on en0: 10.30.175.4.68 > 255.255.255.255.67: BOOTP/DHCP, Request [|bootp] Next I tried a more complex firewall rule: ## Macros local_if="lo0" syn_only="S/FSRA" icmp_types=echoreq ## Tables ## Options ## Scrub scrub in all ## Filter Rules block all anchor "eth/*" # Allow the local interface to talk unrestricted pass in quick on $local_if all pass out quick on $local_if all To this configuration I added rules to the "eth" anchor using pfctl: > cat en0_rules ./en0_rules | pfctl -a eth/en0 -f - Where the en0_rules file contains the following directives: enet_if=en0 syn_only="S/FSRA" icmp_types=echoreq eth_tcp_services="{ 22, 23, 42, 53, 67, 68, 80, 113, 851, 2011, 2021, 6010, 6020, 6667, 7667, 8000, 9402 }" eth_udp_services="{ 22, 23, 42, 53, 67, 68, 851 }" # Default TCP policy block return-rst in log on $enet_if proto TCP all pass in log quick on $enet_if inet proto tcp from any to ($enet_if) port $eth_tcp_services flags $syn_only keep state pass in log quick on $enet_if inet proto udp from any to ($enet_if) port $eth_udp_services # Default ICMP policy block in log on $enet_if proto icmp all pass in log quick on $enet_if inet proto icmp all icmp-type $icmp_types keep state # Allow all outbound traffic block out log on $enet_if all pass out log quick on $enet_if all keep state Notice that both TCP/UDP ports 67,68 are opened. As I understand it, I should only need to allow UDP data out to the server on port 67 and the server should reply back to the client (my target) on UDP port 68. When I tcpdump /dev/pflog0 I see the request go out the interface (en0) but nothing passes back through the firewall (or is logged as being blocked). Will the returning UDP packet on port 68 show up in the pflog0 trace? I don't know all the details of the DHCP handshake, but it would seem the reply back from the server might not contain a destination IP address (since one hasn't been assigned to my target yet) and is routed via the MAC address to my target. Could this be the source of my problems and is there any way to configure a rule to allow the response back from the DHCP server through? So it boils down to a simple question: can anyone suggest a filter rule that will explicitly allow DHCP traffic (only) through an otherwise blocked firewall. Thanks . . . Glenn Schmottlach 2010-11-09T00:09:04Z http://community.qnx.com/sf/go/post73761 Hi everyone, I'm using RTI DDS middleware on the QNX machine which requires UDPv4 installed. I checked that we just had UDPv6 installed on it. How to install UDPv4? Thanks. Yuan Sun 2010-11-05T21:00:29Z http://community.qnx.com/sf/go/post73657 What is the best way to get an io-pkt packet filter ‘unmounted’? I realize there’s no umount support for io-pkt and the suggestion on the wiki is to include a resource manager interface. If I do add a resource manager interface, will something similar to deinit_module in the wiki example really remove the filter from io-pkt (pidin –P io-pkt-v4 mem does not list the filter)? If deinit_module does not completely remove the filter, what are our alternatives? I guess we could start a second instance of io-pkt, add our filter to that second instance and then slay the second io‑pkt instance to clean-up our filter. Any insights are greatly appreciated. Mark Mark Dowdy 2010-11-05T00:32:55Z http://community.qnx.com/sf/go/post73295 I've found that I can detect the alias address and delete it using ifconfig en0 delete 169.254.20.20 and I no longer have the alias (169.... is an example IP alias address). However, the alias comes back after a while. Who is generating the alias -- the auto-ip driver or the dhcp.client? gary anderson 2010-11-02T17:15:02Z http://community.qnx.com/sf/go/post73197 If I use autoIP together with dhcp.client and a DHCP server is able to provide an IP address I also get an alias address from autoIP. I end up with two accessable IP addresses. Is there any way to run dhcp.client so I don't end up with the autoIP alias? This is on a 6.4.1 system. gary anderson 2010-11-01T21:19:04Z