[BlueOnyx:07169] Re: cced gone wild

[Mon Chan]

I try this problem two time.

First time is cmuimport account, the cced is hang up, after i kill the cced
processes, the import is keep going.
And Today the httpd is hang, cannot restart the service,so we need reboot
the server, the cced shutdown very slow.

After reboot the server, all service back to normal. Any solution to fix?


Can i ture off the pam_abl?

[root at ns1 ~]# /etc/rc.d/init.d/pam_abl stop
Clearing pam_abl databases: [  OK  ]
On Sat, Apr 16, 2011 at 4:51 PM, Jeff Jones <jeffrhysjones at mac.com> wrote:





> Hi all,
>
> Just had another occurrence of this mysterious cced hang problem (as
> documented in my early mail here) this morning - this time solved it without
> a hardware reset.
>
> First check running processes for httpd and cced
>
> ps ax | grep http | awk '{print $1}'
> ps ax | grep cced | awk '{print $1}'
>
> What's interesting with this is that there are only a few cced processes
> running, although a large number of httpd.
>
> So first I just try to kill all the http ones in one go:
>
> ps ax | grep http | awk '{print $1}' | xargs kill
>
> That kills all the httpd. However I am still unable to stop cced.
>
> I now have to go:
>
> kill -9 (cced pid)
>
> That kills cced dead.
>
> Now it's just a case of:
>
> Start ceed.init
> Restart admin server
> Start httpd
>
> Done - server back and processes now dropped from 150 back to 100.
>
> So now I think this might be a case of httpd hanging cced - rather than the
> other way round?
>
> Perhaps a 'bad app' ?
>
> No idea why it appears to occur so randomly thought..... the last event was
> 1 month ago....
>
> Cheers,
>
> Jeff
>
>
>
>
> On 16 Mar 2011, at 22:40, Jeff Jones wrote:
>
> > Hi Rashid,
> >
> > I think this may be a similar issue to mine.
> >
> > For me - the following occurs:
> >
> > Steady, linear rise in new processes created on the server, 2 every 5
> minutes
> > Admin GUI stops working
> > Restarting AdmSrv doesn't get GUI working
> > Websites continue to work until RAM on server is exhausted
> > Unable to restart HTTPD
> > Unable to restart cced
> > Unable to restart server
> >
> > The only way to recover is a physical power cycle.
> >
> > I only have one BX server that has this problem, and it occurs about once
> every 6 months.
> >
> > I have set up an SNMP monitoring alert on system processes that alerts me
> when the number of processes rises above normal - in this way I can manually
> intervene and restart the box at night in a 'controlled' manner.
> >
> > But I would really like some tips on how to investigate this problem
> further in order to find the culprit.....
> >
> > Cheers,
> >
> > Jeff
> >
> > On 16 Mar 2011, at 22:18, MuntadaNet Webmaster wrote:
> >
> >> On one of my servers, cced keeps spawning more and more
> >> processes.  Can someone give me an idea of what to look for to
> >> prevent this?  It is causing a Denial of Service eventually in which
> >> the only recovery is cycling the power.
> >>
> >> Regards,
> >>
> >> Rashid
> >>
> >> _______________________________________________
> >> Blueonyx mailing list
> >> Blueonyx at blueonyx.it
> >> http://www.blueonyx.it/mailman/listinfo/blueonyx
> >
> > _______________________________________________
> > Blueonyx mailing list
> > Blueonyx at blueonyx.it
> > http://www.blueonyx.it/mailman/listinfo/blueonyx
>
> _______________________________________________
> Blueonyx mailing list
> Blueonyx at mail.blueonyx.it
> http://mail.blueonyx.it/mailman/listinfo/blueonyx
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.blueonyx.it/pipermail/blueonyx/attachments/20110426/b4237404/attachment.html>