[Imc-web] Re: [IMC-Tech] Tue Jan 17 Apache stop/start

[daniel]

On 1/18/06, Zachary C. Miller <zach at chambana.net> wrote:
> We have these spikes constantly. This has been discussed in the past
> on the imc-tech email list. No one has had the time to spend the
> effort to debug them. The efforts that have been spent have been
> inconclusive. Apache processes get stuck in some kind of infinite busy
> wait loop BEFORE they write a log entry so it is hard to figure out
> what is causing the problem. I believe it to be a problem with our
> version of Dada.
>
> The only thing keeping the server alive is an hourly restart of apache
> by cron. This is ludicrous.
>
agreed

> There are two things that need to be done anyway which may or may not
> solve the problem:
>
> 1) Transition the whole server to apache2.
>
would at the least help to narrow down besides something that will
need to be done eventually anyway...

> 2) Upgrade ALL our dada installs to the latest version of Dada which
> was released late in 2005.
>
i have not familiar with 'dada'... a quick google search produced DaDa Mail?

> (1) takes a lot of time because we have a ton of vhosts that will all
> need to be tested.
>
i have time

> (2) takes a lot of time because we have a ton of local patches to our
> dada code that no one has kept track of.
>
will these need to be ported to the new code?

> I am still looking for someone to volunteer to take on one or both of
> these tasks. I will gladly provide oversight, pointers, and help to
> whoever takes it on but it has to be someone with decent linux admin,
> apache admin, and php experience.
>
i have a bit of experience in all three...

> One easy way to find out if the spikes are caused by dada would be to
> temporarily disable dada for a few days and see if the spikes go
> away. This would also mean the IMC part of the website would be down
> but it might be worth it for the data we'd gather. I think I proposed
seeing that wrfu is running a fairly simple install of drupal it would
not be much to replicate the db (assuming its using mysql) and run it
on one of my servers during this period.  a simple a record change or
meta redirect on the server would do for that time.  Although i
wouldn't think it would come to that *hopefully*.

> this a while ago but dropped the ball on following up with it because
> of the chaos of the holidays.
i hear you on the fun of holidays...
>
i am planning on attending the stearing meeting at 8pm on thursdy as
well as the meeting at 7pm.  time permiting, if you would like to
discuss the matter further and possibly access for myself to the
server at that point (provided my 'bit of experience' seems fair
enough).

> daniel wrote:
> > where there any access errors or server errors that would lead to the
> > cause of the spikes?
> > On 1/17/06, Daniel Lewart <lewart at uiuc.edu> wrote:
> > > Zach, et al,
> > >
> > > Twice today I had to stop Apache because it was sucking up
> > > CPU and then restart it.
> > >
> > > According to /var/log/uptime.log, at 16:41:39 the Load Average was 62.
> > > When it gets this high, the mail queues effectively stop.
> > >
> > > Cheers,
> > > Dan
> > > _______________________________________________
> > > IMC-Tech mailing list
> > > IMC-Tech at lists.ucimc.org
> > > http://lists.chambana.net/cgi-bin/listinfo/imc-tech
> > >
> >
> >
> > --
> > Thanks,
> > Daniel Meredith
> > Professional Geek
> > _______________________
> > Champaign, IL 61820
> >
> > 217.714.2738 (ph)
> > dan.blah at gmail.com
> > _______________________________________________
> > IMC-Tech mailing list
> > IMC-Tech at lists.ucimc.org
> > http://lists.chambana.net/cgi-bin/listinfo/imc-tech
> >
>
> --
> Zachary C. Miller - @= - http://zach.chambana.net/
> IMSA 1995 - UIUC 2000 - Just Another Leftist Muppet - Ya Basta!
>  Social Justice, Community, Nonviolence, Decentralization, Feminism,
>  Sustainability, Responsibility, Diversity, Democracy, Ecology
>


--
Thanks,
Daniel Meredith
Professional Geek
_______________________
Champaign, IL 61820

217.714.2738 (ph)
dan.blah at gmail.com