This problem has been resolved.  In case anyone else sees a similar
problem, the solution was to build a custom kernel with 2 changes from
the stock kernel provided with the Linux distribution:

1.  Processor family changed from default "Pentium Pro" to "AMD Athlon
64 (K8)" as we are running on a quad-core AMD Phenom CPU.
2.  Changed the Preemption model from the default 'Desktop' to
'Server.'

Since booting this new kernel (same 2.6.33.4-smp version), we've had
no more exceptions thrown when Python parent processes communicate
with their children.

Thanks,
Dennis


On Oct 28, 3:53 pm, Dennis <dennisfraz...@charter.net> wrote:
> Thanks to both for the replies.  We're using Slackware 13.1, 32-bit.
>
> uname -a output:
>
> Linux svrname 2.6.33.4-smp #2 SMP Wed May 12 22:47:36 CDT 2010 i686
> AMD Phenom(tm) 9950 Quad-Core Processor AuthenticAMD GNU/Linux
> r...@qeweb1:/usr/lib/python2.6#
>
> How do we check for SELinux?
>
> What is required to try mod_wsgi?
>
> Thanks,Dennis
>
> On Oct 28, 3:01 pm, Christian Hammond <chip...@chipx86.com> wrote:
>
>
>
>
>
>
>
> > I'm curious why people hit this problem. We use mod_python on
> > reviews.reviewboard.org and are aware of its use in lots of large
> > installations, and this problem hasn't come up more than a couple times.
>
> > What version of which distro are you using? By any chance is SELinux
> > enabled?
>
> > Maybe give mod_wsgi a shot?
>
> > Christian
>
> > --
> > Christian Hammond - chip...@chipx86.com
> > Review Board -http://www.reviewboard.org
> > VMware, Inc. -http://www.vmware.com
>
> > On Thu, Oct 28, 2010 at 11:39 AM, Rob <rrom...@gmail.com> wrote:
> > > See also topic
> > >http://groups.google.com/group/reviewboard/browse_thread/thread/af0e3...
>
> > > I have seen this problem with every version of ReviewBoard prior to
> > > 1.5. I am stuck at 1.0.4 at the moment.
>
> > > What I did to fix this was modify all the calls in the diffiviewer
> > > code below to use os.system() instead of the Python subprocess module.
> > > Something about the subprocess module does NOT play nicely under
> > > mod_python/Apache.
>
> > > Now for 1.5, I cannot upgrade because post-review is triggering a call
> > > to the new Paramiko module, which calls into Python's 'platform'
> > > module for the system() call.  It turns out platform.system() ALSO
> > > uses subprocess.Popen instead of os.system().
>
> > > I'm glad I'm not the only one seeing this. I bet it's a mod_python
> > > bug, Incidentally, mod_python is now a dead project.
>
> > > On Oct 27, 10:51 am,Dennis<dennisfraz...@charter.net> wrote:
> > > > Using ReviewBoard 1.5 on Linux with Apache 2.2.15 and mod_python, we
> > > > are getting intermittent but fairly frequent errors when attempting to
> > > > view diffs with Perforce:
>
> > > > (10, 'No child processes')
>
> > > > Traceback (most recent call last):
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/views.py", line 153, in view_diff
> > > >     interdiffset, highlighting, True)
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/diffutils.py", line 1063, in get_diff_files
> > > >     large_data=True)
> > > >   File "/usr/lib/python2.6/site-packages/Djblets-0.6.4-py2.6.egg/
> > > > djblets/util/misc.py", line 166, in cache_memoize
> > > >     data = lookup_callable()
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/diffutils.py", line 1062, in <lambda>
> > > >     enable_syntax_highlighting)),
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/diffutils.py", line 548, in get_chunks
> > > >     old = get_original_file(filediff)
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/diffutils.py", line 364, in get_original_file
> > > >     large_data=True)[0]
> > > >   File "/usr/lib/python2.6/site-packages/Djblets-0.6.4-py2.6.egg/
> > > > djblets/util/misc.py", line 166, in cache_memoize
> > > >     data = lookup_callable()
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/diffutils.py", line 363, in <lambda>
> > > >     data = cache_memoize(key, lambda: [fetch_file(file, revision)],
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/diffviewer/diffutils.py", line 342, in fetch_file
> > > >     data = tool.get_file(file, revision)
> > > >   File "/usr/lib/python2.6/site-packages/ReviewBoard-1.5-py2.6.egg/
> > > > reviewboard/scmtools/perforce.py", line 93, in get_file
> > > >     (res, errdata) = p.communicate()
> > > >   File "/usr/lib/python2.6/subprocess.py", line 689, in communicate
> > > >     return self._communicate(input)
> > > >   File "/usr/lib/python2.6/subprocess.py", line 1233, in _communicate
> > > >     self.wait()
> > > >   File "/usr/lib/python2.6/subprocess.py", line 1157, in wait
> > > >     pid, sts = os.waitpid(self.pid, 0)
> > > > OSError: [Errno 10] No child processes
>
> > > > Often, repeating the attempt 2-3 times will eventually result in
> > > > success.  Any ideas on root cause or a workaround?
>
> > > > Thanks very much,
> > > >Dennis
>
> > > --
> > > Want to help the Review Board project? Donate today at
> > >http://www.reviewboard.org/donate/
> > > Happy user? Let us know athttp://www.reviewboard.org/users/
> > > -~----------~----~----~----~------~----~------~--~---
> > > To unsubscribe from this group, send email to
> > > reviewboard+unsubscr...@googlegroups.com<reviewboard%2bunsubscr...@googlegr
> > >  ­oups.com>
> > > For more options, visit this group at
> > >http://groups.google.com/group/reviewboard?hl=en-Hide quoted text -
>
> > - Show quoted text -

-- 
Want to help the Review Board project? Donate today at 
http://www.reviewboard.org/donate/
Happy user? Let us know at http://www.reviewboard.org/users/
-~----------~----~----~----~------~----~------~--~---
To unsubscribe from this group, send email to 
reviewboard+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/reviewboard?hl=en

Reply via email to