[mod_python] Apache crashing

Dan Eloff dan.eloff at gmail.com
Sat Feb 11 17:44:05 EST 2006


Hmm that sounds like it fits the bill. It looks like an evil issue though,
perhaps the way to fix it is not just through mod_python, but also by
improving how apache handles the situation? I don't understand things at
that level enough to be of any help, I'm sure you guys will do the right
thing though with Graham on the case.

-Dan

On 2/11/06, Graham Dumpleton <grahamd at dscpl.com.au> wrote:
>
>
> On 12/02/2006, at 9:05 AM, Dan Eloff wrote:
>
> > I've been using mod_python (3.2.2b the last version released for
> > win) for a while now, and I'm really fond of the flexibility it
> > gives me, but I've run across a problem as of late.
> >
> > Apache crashes when shut down with this error, apparently while
> > waiting for worker threads to shut down. The CPU pins up at 100%
> > and then after a few seconds apache crashes.
> >
> > [Sat Feb 11 13:33:56 2006] [crit] [Sat Feb 11 13:33:56 2006] file C:
> > \\asf-build\\build-2.0.55\\server\\mpm\\winnt\\child.c, line 1078,
> > assertion "(rv >= 0) && (rv < threads_created)" failed
> >
> > Has anybody had trouble with this?
>
> It may be related to:
>
>    http://issues.apache.org/jira/browse/MODPYTHON-109
>
> Problem is that mod_python tries to call Py_Finalize() when child
> process
> being killed off, but it is being done from inside signal handler
> context which
> is quite bad. Similar problem if you use req.server.register_cleanup
> () or
> apache.register_cleanup(). The latter functions may have to be removed
> if no equivalent can be found that is safe.
>
> Graham
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mm_cfg_has_not_been_edited_to_set_host_domains/pipermail/mod_python/attachments/20060211/55053c78/attachment.html


More information about the Mod_python mailing list