[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

panics on 24 hour boundaries

Actually, one note on this, I had a server that was panicing similar to 
this that DES was kind enough to look at a while ago.  The problem was 
never tracked down, but I found that taking INET6 out of the kernel solved 
the problem.  I asked michael_(_at_)_gargantuan_(_dot_)_com if this was possible to try, 
and he said they make heavy use of INET6 so they could not take it out. The 
other user seeing similar crashes (tss_(_at_)_reflection_(_dot_)_co_(_dot_)_jp) also makes use of 
INET6. It could of course be total coincidence and have nothing to do with it.


At 03:03 PM 01/10/2003, Bill Vermillion wrote:
>On Wed, Oct 01, 2003 at 11:05 , Men gasped, women
>fainted, and small children were reduced to tears as
>freebsd-stable-request_(_at_)_freebsd_(_dot_)_org confessed to all:
> > Message: 18
> > Date: Tue, 30 Sep 2003 23:19:34 -0400
> > From: "Michael W. Oliver" <michael_(_at_)_gargantuan_(_dot_)_com>
> > Subject: Re: panics on 24 hour boundaries
>[severly edited - wjv]
> > +--- On Tuesday, September 30, 2003 22:35 ---
> > | Robert Watson proclaimed:
> > | Initial reactions: panics on 24 hour boundaries are, in my
> > | experience, often associated with the daily event. Once a
> > | day, the daily scripts run find several times on your file
> > | systems, causing every file and directory to be inspected
> > | for changes in setuid scripts, etc. This can trigger certain
> > | classes of race conditions and resource limits that you might
> > | otherwise not hit in normal operation -- and conviently, they
> > | run 24 hours apart :-). To try and confirm this suspicion,
> > | it would be interesting to know what time of day exactly the
> > | panics take place, and whether you can reproduce the panic by
> > | manually running the daily or security script.
> > All of the panics happened in the evening hours, between 1800
> > and 2200 EDT. I am also able to successfully run the daily
> > periodic scripts at any time of the day without issue.
>One part in this thread said the panics happened 24 hours after
>reboot and that would imply something in scripts that depend upon
>a length of time being powered up.
>However if all the pnaics occur in the 1800-2200 time frame
>this could be caused by an external event.
>It could be any large device on the same electric circuits you are
>on.  By the 'same circuit' I mean anyone and/or anything connected
>to the same power transformer.  In a residential area this could be
>several houses.
>Anything that could put a spike on the line could cuase this.
>And even if the computer if filter and on a UPS if any device
>connected to the computer is not also on the same filter those
>could be the culprits.  I've seen [in the far past when I
>maintained many machines with serial terminal] terminals and also
>printers cause this.
>One place had contruction going on next door and that was alway
>in early afternoon when one piece of equipment was fired up.
>And one of the legendary stories is about the systems that paniced
>every day between noon and about 10 after.  That was traced to a
>microwave in the lunchroom.
>It's been my experience that often time related crashes are
>external to the machines involved unless each and everything
>connected to the machine is coming from the same filterer/protected
>source, including all phone lines for DSL and or cable.
> > End of freebsd-stable Digest, Vol 28, Issue 4
>Bill Vermillion - bv @ wjv . com
>freebsd-stable_(_at_)_freebsd_(_dot_)_org mailing list
>To unsubscribe, send any mail to "freebsd-stable-unsubscribe_(_at_)_freebsd_(_dot_)_org"