[mythtv-users] OT. Howto diagnose machine dying/crashing

Mark mark at ripitup.ca
Fri Sep 15 17:17:21 UTC 2006


On Friday 15 September 2006 11:06, Yeechang Lee wrote:
> Mark <mark at ripitup.ca> says:
> > About every week or 10 days, this box just up and dies.  No
> > /var/log/messages entries ( you just see the reboot ), nothing on the
> > screen etc...
>
> [...]
>
> > It does seem to be load related because I have reproduced it once by
> > transfering a hundred or so gigs of data into it and it died then.
> > FC5, mdadm raid..
>
> Other than FC5, this sounds *exactly* like what I faced with an
> eight-drive FC3 system using mdadm RAID 5; see
> <URL:https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=167173>. Was
> the kernel stack issue not resolved in FC5?
>
> The other thing that comes to mind is the power supply. Is it a brand
> name or a $39 wonder? 410W is plenty, but so are four RAID drives plus
> a boot drive and perhaps an optical drive, and cheap power supplies
> are notorious for buckling under strain. That aforementioned RAID
> server is currently out of commission due to random hangs (after a few
> minutes or hours, formerly 10-14 days); I think the (Antec, thus
> quality) 550W drive is responsible, and haven't gotten around to
> replace it with another (Antec) supply yet.
I have actually swapped out the power supply for another I had as well.  Same 
story.
I am looking for a crash dump utility but am having some issues with netdump ( 
cant arp? ) and I dont have a free parition to use with diskdump.  Anyone 
know of any other ways to capture crash data?

Mark



More information about the mythtv-users mailing list