[horde] help. new install, something is broken in imap (solved) but exception is broken..

[LALOT Dominique]

2012/2/4 Vilius Šumskas <vilius at lnk.lt>

> Sveiki,
>
> Saturday, February 4, 2012, 1:06:39 AM, you wrote:
>
> > 2012/2/2 Vilius Šumskas <vilius at lnk.lt>
>
> >
> >> 2012/2/2 Michael M Slusarz <slusarz at horde.org>
>  >>
>  >> > Quoting LALOT Dominique <dom.lalot at gmail.com>:
>  >> >
>  >> >  2012/2/1 Michael M Slusarz <slusarz at horde.org>
>  >> >>
>  >> >>  Quoting LALOT Dominique <dom.lalot at gmail.com>:
>  >> >>>
>  >> >>>  Is there a piece of code I could put somewhere to test and
> isolate the
>  >> >>>
>  >> >>>> problem? This setup is running for me for months and I have no
>  >> problems,
>  >> >>>> except some suspect and unusual traces. I have may be some other
>  >> related
>  >> >>>> problems, that could be the same and are segfaulting apache:
>  >> >>>>
>  >> >>>> [Wed Feb 01 12:59:17 2012] [error] [client xx] PHP Fatal error:
>  Call
>  >> >>>> to a
>  >> >>>> member function getName() on a non-object in
> >
> >> >>>> /var/www/horde/imp/lib/Ajax/****Application.php on line 1333,
>  >> referer:
>  >> >>>> http://xx/horde/imp/message-
>  >> ****dimp.php?ajaxui=1&mailbox=**<http://xx/horde/imp/message-
>  >> **dimp.php?ajaxui=1&mailbox=**>
>  >> >>>>
>  >> SU5CT1gvZW52b2lz&uid=5&uniq=****1328097524472<http://xx/horde/**
>  >> >>>> imp/message-
>  >> dimp.php?ajaxui=1&**mailbox=SU5CT1gvZW52b2lz&uid=**
>  >> >>>> 5&uniq=1328097524472<http://xx/horde/imp/message-
>  >> dimp.php?ajaxui=1&mailbox=SU5CT1gvZW52b2lz&uid=5&uniq=13280975244
> >
> >> 72>
>  >> >>>> >
>  >> >>>>
>  >> >>>> [Wed Feb 01 12:59:17 2012] [notice] child pid 29166 exit signal
>  >> >>>> Segmentation fault (11), possible coredump in /tmp
>  >> >>>>
>  >> >>>>
>  >> >>> PHP should NEVER coredump.  If it is, there is something seriously
> wrong
>  >> >>> with it or one of its components.
>  >> >>>
>  >> >>> Yes Michael, I known that but each compiled pear module can be a
> cause
>  >> of
>  >> >>>
>  >> >> coredump
>  >> >>
>  >> >> root at horde41:/tmp# lsb_release -a
>  >> >> No LSB modules are available.
>  >> >> Distributor ID: Ubuntu
>  >> >> Description:    Ubuntu 10.10
>  >> >> Release:        10.10
>  >> >> Codename:       maverick
>  >> >>
>  >> >> Just looking at last trace (there is no debugging php symbols in
> that
>  >> >> distrib)
>  >> >>
>  >> >> (gdb) bt
>  >> >> #0  0x00007f64ea623f89 in _zend_hash_add_or_update () from
> >
> >> >> /usr/lib/apache2/modules/**libphp5.so
> >
> >> >>
>  >> >
>  >> > Are you using APC?  Try disabling that.
>  >> >
>  >> >
>  >> > michael
>  >> >
> >
> >> > ______________________________**_____
> >
> >> > Michael Slusarz [slusarz at horde.org]
>  >> >
>  >> >
>  >> I am using xcache, but I tried many options to enable/disable modules
> for
>  >> example, Net_DNS2 is crashing (method not found and segfault..)
>  >> I'm quite sure I was runing without xcache and having same problems.
> >
> >
> > Usual things to check when Apache segfaulting.
> >
> >  1. Check server RAM.
> >  2. Don't run self compilled software or software which comes from
> > 3rd party repositories and extensivelly uses other original software
> > on the machine. For example don't run 3rd party PHP RPMs with original
> Apache.
> >   3. Try to completely uninstall PHP extensions and pecl modules one by
> one.
> >  4. Try to disable Apache modules one by one.
> >
>
>
> > I have several VM, so RAM is OK. I already tried to remove some package
> without sucess.
> > My boss went to shown me some problems he had and I was able to see
> > that we got some segfault when forwarding a mail containing an attach
> file.
> >  I see green sent OK, then 2 red warning Pb communicating with the
> > server. Does it mean something to you?
>
> It   doesn't  matter  if other VMs run OK. You should still try to change
> the
> RAM or at least check it with special software.
>
> --
> Best regards,
>  Vilius
>
> --
> Horde mailing list
> Frequently Asked Questions: http://horde.org/faq/
> To unsubscribe, mail: horde-unsubscribe at lists.horde.org
>

I get this problem on several VM and our Vmware infra is OK. It always
happens in the same part of code. We run in a cluster. So there will be big
problems somewhere else.

Dom

-- 
Dominique LALOT
Ingénieur Systèmes et Réseaux
http://annuaire.univmed.fr/showuser.php?uid=lalot