First thing which comes to mind is /dev/shmem. Any space used by files in there will count towards procnto's memory 
usage. What's in /dev/shmem when procnto is using 260MiB?

Sent from my BlackBerry 10 smartphone on the Rogers network.
  Original Message
From: Davide Ancri
Sent: Wednesday, November 1, 2017 6:45 AM
To: ostech-core_os
Reply To: ostech-core_os@community.qnx.com
Subject: [QNX6.5] procnto using 260MB of data and system instability


hi all

QNX6.5 running on a x86 platform: we are facing instability problems, showing different symptoms:
- complete system irresponsiveness:  no network activity, no local console reaction
- network no more responsive, local console available but unable to run new processes (getting stuck when trying to)

We have a special system memory distribution, so that the qnx kernel runs with a limited amount of ram physically below 
4 GB, plus a good amount of ram above 4 GB (see the "pidin syspage" asinfo sections into the attachment).

During system run, a lot of processes start (many of them enabled to >4GB phys ram, with "on -P"), and what I can see 
just before the system gets stuck is "pidin mem" reporting procnto making use of a lot of ram for its data (260 MB, more
 than than the total available sysram phys <4GB).

Can this big amount of procnto data be related to system instability? Or can it be symptom of other problems (too many 
processes? lack of realtime at user-lever priorities?)

Can anyone explain which are the possible reasons for pushing procnto to allocate lot of data?

The attachment shows output from "pidin syspage" (asinfo section) and "pidin mem", giving an overview on amount of 
running processes and memory utilization.
Don't hesistate to ask for other infos.

thanks a lot for any support,
Davide




_______________________________________________

OSTech
http://community.qnx.com/sf/go/post118166
To cancel your subscription to this discussion, please e-mail ostech-core_os-unsubscribe@community.qnx.com