How can I unblocked the process from io-net or kill my process without
having to terminate io-net process or shutdown? Is this a known problem with
io-net?
It’s not a known problem. Can you be more specific?
-seanb
JS <jsukamtoh@infolink.co.id> wrote:
How can I unblocked the process from io-net or kill my process without
having to terminate io-net process or shutdown? Is this a known problem with
io-net?
Please post the pidin output.
-seanb
JS <jsukamtoh@infolink.co.id> wrote:
We are using Neutrino 6.2.1. The process consists of 2 threads, both are
interfacing to the same serial port, one is receiving data from the serial
port and pass it to another process via socket, whereas the other thread is
receiving data from another process via socket and write it to the serial
port. At one state, this process is trying to terminate itself but it hang.
One thread is waiting for close to return, while the other is waiting for
write to socket to return (we use debugger to check its function call).
pidin -p shows that both threads are blocked by io-net process, and we can’t
kill or do anything to the process. Other processes are still running and so
is the io-net. This condition is not reproduceable yet. It happens only once
but quite critical because we have to shutdown the operating system in order
to kill the process to release the serial port.“Sean Boudreau” <> seanb@qnx.com> > wrote in message
news:db3284$sub$> 1@inn.qnx.com> …It’s not a known problem. Can you be more specific?
-seanb
JS <> jsukamtoh@infolink.co.id> > wrote:
How can I unblocked the process from io-net or kill my process without
having to terminate io-net process or shutdown? Is this a known problem
with
io-net?
We are using Neutrino 6.2.1. The process consists of 2 threads, both are
interfacing to the same serial port, one is receiving data from the serial
port and pass it to another process via socket, whereas the other thread is
receiving data from another process via socket and write it to the serial
port. At one state, this process is trying to terminate itself but it hang.
One thread is waiting for close to return, while the other is waiting for
write to socket to return (we use debugger to check its function call).
pidin -p shows that both threads are blocked by io-net process, and we can’t
kill or do anything to the process. Other processes are still running and so
is the io-net. This condition is not reproduceable yet. It happens only once
but quite critical because we have to shutdown the operating system in order
to kill the process to release the serial port.
“Sean Boudreau” <seanb@qnx.com> wrote in message
news:db3284$sub$1@inn.qnx.com…
It’s not a known problem. Can you be more specific?
-seanb
JS <> jsukamtoh@infolink.co.id> > wrote:
How can I unblocked the process from io-net or kill my process without
having to terminate io-net process or shutdown? Is this a known problem
with
io-net?
I don’t have it anymore because it occurred 12 hours ago and I have to
reboot the system. I’ll post it again if I could reproduce the problem. What
else do you need besides pidin output?
“Sean Boudreau” <seanb@qnx.com> wrote in message
news:db3alh$5t9$1@inn.qnx.com…
Please post the pidin output.
-seanb
JS <> jsukamtoh@infolink.co.id> > wrote:
We are using Neutrino 6.2.1. The process consists of 2 threads, both are
interfacing to the same serial port, one is receiving data from the
serial
port and pass it to another process via socket, whereas the other thread
is
receiving data from another process via socket and write it to the
serial
port. At one state, this process is trying to terminate itself but it
hang.
One thread is waiting for close to return, while the other is waiting
for
write to socket to return (we use debugger to check its function call).
pidin -p shows that both threads are blocked by io-net process, and we
can’t
kill or do anything to the process. Other processes are still running
and so
is the io-net. This condition is not reproduceable yet. It happens only
once
but quite critical because we have to shutdown the operating system in
order
to kill the process to release the serial port.“Sean Boudreau” <> seanb@qnx.com> > wrote in message
news:db3284$sub$> 1@inn.qnx.com> …It’s not a known problem. Can you be more specific?
-seanb
JS <> jsukamtoh@infolink.co.id> > wrote:
How can I unblocked the process from io-net or kill my process
without
having to terminate io-net process or shutdown? Is this a known
problem
with
io-net?
\
On Wed, 13 Jul 2005 12:42:44 +0000, Sean Boudreau wrote:
It’s not a known problem. Can you be more specific?
There was definitely a problem with tasks beeing
NET-REPLY blocked, that couldn’t be unblocked by
anything (signal, TimerTimout, etc.).
It was solved in the meantime by a SP and it was also
documented in the release notes, IIRC.
If anybody is interested, I could checkout the PR number,
as soon as I’m back in the office 
–
Karsten P. Hoffmann <karsten.p.hoffmann@web.de>
“I love deadlines. I especially like the whooshing sound
they make as they go flying by.”
[In memoriam Douglas Adams, 1952-2001]
Great. We look forward to that.
“Karsten P. Hoffmann” <karsten.p.hoffmann@web.de> wrote in message
news:pan.2005.07.20.22.17.34.25442@web.de…
On Wed, 13 Jul 2005 12:42:44 +0000, Sean Boudreau wrote:
It’s not a known problem. Can you be more specific?
There was definitely a problem with tasks beeing
NET-REPLY blocked, that couldn’t be unblocked by
anything (signal, TimerTimout, etc.).It was solved in the meantime by a SP and it was also
documented in the release notes, IIRC.If anybody is interested, I could checkout the PR number,
as soon as I’m back in the office >–
Karsten P. Hoffmann <> karsten.p.hoffmann@web.de
“I love deadlines. I especially like the whooshing sound
they make as they go flying by.”
[In memoriam Douglas Adams, 1952-2001]
Is this problem fixed in 6.3? Thanks.
“Karsten P. Hoffmann” <karsten.p.hoffmann@web.de> wrote in message
news:pan.2005.07.20.22.17.34.25442@web.de…
On Wed, 13 Jul 2005 12:42:44 +0000, Sean Boudreau wrote:
It’s not a known problem. Can you be more specific?
There was definitely a problem with tasks beeing
NET-REPLY blocked, that couldn’t be unblocked by
anything (signal, TimerTimout, etc.).It was solved in the meantime by a SP and it was also
documented in the release notes, IIRC.If anybody is interested, I could checkout the PR number,
as soon as I’m back in the office >–
Karsten P. Hoffmann <> karsten.p.hoffmann@web.de
“I love deadlines. I especially like the whooshing sound
they make as they go flying by.”
[In memoriam Douglas Adams, 1952-2001]
On Fri, 22 Jul 2005 19:17:33 +0700, JS wrote:
Is this problem fixed in 6.3? Thanks.
Yes, it was fixed in 6.3, not in 6.2 …
–
Karsten P. Hoffmann <karsten.p.hoffmann@web.de>
“I love deadlines. I especially like the whooshing sound
they make as they go flying by.”
[In memoriam Douglas Adams, 1952-2001]
Dear Karsten,
Where can I get the SP? Thanks.
JS
“Karsten P. Hoffmann” <karsten.p.hoffmann@web.de> wrote in message
news:pan.2005.07.27.21.29.13.486987@web.de…
On Fri, 22 Jul 2005 19:17:33 +0700, JS wrote:
Is this problem fixed in 6.3? Thanks.
Yes, it was fixed in 6.3, not in 6.2 …
–
Karsten P. Hoffmann <> karsten.p.hoffmann@web.de
“I love deadlines. I especially like the whooshing sound
they make as they go flying by.”
[In memoriam Douglas Adams, 1952-2001]
On Fri, 12 Aug 2005 18:32:05 +0700, JS wrote:
Where can I get the SP? Thanks.
I guess, you hav to contact you QNX dealer.
Perhaps you can download it, if you register your current
version in ‘MyQNX’ and have a look into the download area,
then … but I’m not sure about this.
Sorry.
–
Karsten P. Hoffmann <karsten.p.hoffmann@web.de>
“I love deadlines. I especially like the whooshing sound
they make as they go flying by.”
[In memoriam Douglas Adams, 1952-2001]