Re: Oracle 10g hangs intermittently waiting for I/O
Date: Sun, 17 May 2009 21:32:10 +0200
Yes, all volumes are RAID-5 sets. I am now working with the storage guys and hope that we will be able to thoroughly check all the subsystems. I will share our findings to the list.
Thanks for all suggestions!
On 2009-05-16 22:02, Cary Millsap wrote:
> I keep thinking "RAID cache flush" as I read this thread, but I don't
> think I've seen anyone mention it yet. RAID 5 perchance? I hate guessing
> instead of measuring...
> Cary Millsap
> Method R Corporation
> 2009/5/16 Rajeev Prabhakar <rprabha01_at_gmail.com
> Hi Pawel,
> In no particular order, please check with your sysadmins/storage admins
> about the following :
> a) Queue depth of the LUNs / associated optimal settings for your
> b) Any unsuccessful microcode/firmware upgrade on the FC fabric.
> c) Double check for any combination of any FC/non-FC drives
> OR any failure of drives/associated raid group rebuild errors.
> d) LUN level i/o statistics and any out of ordinary observations.
> 2009/5/16 Paweł Kotlarz <pkotla_at_go2.pl>:
> > Thanks, Rajeev. I appreciate your help very much!
> > Pawel