Project Home
Project Home
Wiki
Wiki
Discussion Forums
Discussions
Project Information
Project Info
Forum Topic - inetd in SIGSUSPEND state: (7 Items)
   
inetd in SIGSUSPEND state  
I'm investigating a problem we're having telnet'ing into a few boards in
our system (we're getting Connection refused). We can successfully
telnet to one of the three boards in our system but not the other two.
inetd seems to be running on all three boards (it's in the pidin output)
but the state of inetd on the two boards we're having problems with is
SIGSUSPEND (the state of inetd on the working board is SIGWAITINFO). Can
anyone explain what it means to have inetd in SIGSUSPEND state how it
can get this way? Is there a way to prod inetd out of this state?

 

While searching for information on this issue I found mention on running
inetd with the -d switch to get debug information. Unfortunately, our
two problem boards are booting off compact flash and I'm trying to
figure out how inetd is getting started at all (there's not an
/etc/rc.d/rc.local file in the compact flash file system).

 

Thanks.

 

Mark

Re: inetd in SIGSUSPEND state  
On Fri, Jul 31, 2009 at 09:16:19PM -0400, Mark Dowdy wrote:
> I'm investigating a problem we're having telnet'ing into a few boards in
> our system (we're getting Connection refused). We can successfully
> telnet to one of the three boards in our system but not the other two.
> inetd seems to be running on all three boards (it's in the pidin output)
> but the state of inetd on the two boards we're having problems with is
> SIGSUSPEND (the state of inetd on the working board is SIGWAITINFO). Can
> anyone explain what it means to have inetd in SIGSUSPEND state how it
> can get this way? Is there a way to prod inetd out of this state?
> 
>  
> 
> While searching for information on this issue I found mention on running
> inetd with the -d switch to get debug information. Unfortunately, our
> two problem boards are booting off compact flash and I'm trying to
> figure out how inetd is getting started at all (there's not an
> /etc/rc.d/rc.local file in the compact flash file system).
> 

This is 6.3 right?

inetd was probably:
- started with no /etc/inet.conf present
- started with no services present / uncommented in /etc/inetd.conf
- started as non root and wasn't able to bind to the port(s)
  required for the services specified in /etc/inetd.conf

Regards,

-seanb
Re: inetd in SIGSUSPEND state  
Correct, 6.3.

/etc/inet.conf is present on the non-working cards but it is a link to /etc/inet.conf on the board on which inetd is 
working. I'll have to do some more investigation into our start-up sequence to make sure the link exists before inetd 
starts on CF cards.

Thanks for the lead.

Mark
Re: inetd in SIGSUSPEND state  
OK, I collected the debug information for inetd but nothing jumps out at me as the reason why this daemon would be stuck
 in SIGSUSPEND. Here's the debug output.

ADD : ftp proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/ftpd policy=
ADD : telnet proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
ADD : shell proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/rshd policy=
ADD : telnet proto=tcp6, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
ADD : pidin proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
ADD : pidinmem proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
ADD : pidinarg proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
ADD : pidinenv proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
ADD : pidinsig proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
ADD : phrelay proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/bin/phrelay policy=
ADD : discard proto=tcp, wait:max=0.40, user:group=root. builtin=80516b0 server=internal policy=
ADD : discard proto=udp, wait:max=0.40, user:group=root. builtin=80516c0 server=internal policy=
FREE: discard proto=udp, wait:max=0.40, user:group=root. builtin=80516c0 server=internal policy=
FREE: discard proto=tcp, wait:max=0.40, user:group=root. builtin=80516b0 server=internal policy=
FREE: phrelay proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/bin/phrelay policy=
FREE: pidinsig proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
FREE: pidinenv proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
FREE: pidinarg proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
FREE: pidinmem proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
FREE: pidin proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
FREE: telnet proto=tcp6, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
FREE: shell proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/rshd policy=
FREE: telnet proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
FREE: ftp proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/ftpd policy=

Any ideas? Thanks.
Re: inetd in SIGSUSPEND state  
On Tue, Aug 04, 2009 at 05:20:32PM -0400, Mark Dowdy wrote:
> OK, I collected the debug information for inetd but nothing jumps out at me as the reason why this daemon would be 
stuck in SIGSUSPEND. Here's the debug output.
> 
> ADD : ftp proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/ftpd policy=
> ADD : telnet proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
> ADD : shell proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/rshd policy=
> ADD : telnet proto=tcp6, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
> ADD : pidin proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> ADD : pidinmem proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> ADD : pidinarg proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> ADD : pidinenv proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> ADD : pidinsig proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> ADD : phrelay proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/bin/phrelay policy=
> ADD : discard proto=tcp, wait:max=0.40, user:group=root. builtin=80516b0 server=internal policy=
> ADD : discard proto=udp, wait:max=0.40, user:group=root. builtin=80516c0 server=internal policy=
> FREE: discard proto=udp, wait:max=0.40, user:group=root. builtin=80516c0 server=internal policy=
> FREE: discard proto=tcp, wait:max=0.40, user:group=root. builtin=80516b0 server=internal policy=
> FREE: phrelay proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/bin/phrelay policy=
> FREE: pidinsig proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> FREE: pidinenv proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> FREE: pidinarg proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> FREE: pidinmem proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> FREE: pidin proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/bin/pidin policy=
> FREE: telnet proto=tcp6, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
> FREE: shell proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/rshd policy=
> FREE: telnet proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/telnetd policy=
> FREE: ftp proto=tcp, wait:max=0.40, user:group=root. builtin=0 server=/usr/sbin/ftpd policy=
> 
> Any ideas? Thanks.

So it wasn't able to register any services.  Is the
stack up and running prior to starting inetd?  There
may be extra stuff logged to the syslog in this case:
set up /etc/syslog.conf and run syslogd.

-seanb
Re: inetd in SIGSUSPEND state  
Doh. The stack was not started. Difficult to get network services without the stack. Thanks for deciphering the log 
messages.
RE: inetd in SIGSUSPEND state  
> The stack was not started

Don't feel bad - some of the error messages
are not that clear.

For example, simply run 

  # ifconfig

without the stack running, and it will respond
with some odd error about your family.

--
aboyd