[wellylug] High load averages but no apparent cause

[Daniel Reurich]

On Thu, 2010-03-25 at 10:31 +1100, Daniel Pittman wrote:
> David Harrison <david.harrison at stress-free.co.nz> writes:
> 
> > No, but now that you say that if the system is unable to write to the RAID5
> > which contains the log file would this even happen?
> >
> > e.g. /var is the problematic RAID5 partition and when it locks up it takes
> > out one or more of the physical disks.
> >
> > An interesting observation is that when the problem occurs it either locks
> > up both sda & sdb, or sdc by itself.  I am guessing that this is because sda
> > & sdb are on the same channel, so either the channel itself is going or one
> > of the disks is which is taking the other with it.
> 
> That is extremely unlikely: SATA disks don't have the older PATA "shared
> channel" issue, and as far as I can tell these are SATA disks, right?
> 
Maybe.  It may well be the the phy's that drive the signals down the
sata cables typically do pairs of channels, but I'm only speculating
here.

I have seen this behaviour before on 2 servers both with 4 disks where
the power supply wasn't quire reliable enough anymore, and it would
periodically drop a disk or 2 for a few seconds under load.


-- 
Daniel Reurich.

Centurion Computer Technology (2005) Ltd
Mobile 021 797 722