From: John Malick (john@starinc.com)
Date: Thu Apr 07 2005 - 10:48:08 EDT
I forgot to mention that I'm running kernel patch 108528-29 on these
servers.
I have found that 108528-28 apparently does indeed cause this problem
when running Veritas VxFS.
The solution, at the time, seemed to have been to backout of -28 to at
least 108528-27. Again, notice that we are at rev -29. We cannot backout
to -27 for other reasons (apps require -28 or higher).
Does anyone know if -29 of the kernel patch have the same issues? Just
to remind everyone that I have two identical systems and one is not
showing the problem.
Thanks again
John Malick wrote:
> Running Solaris 8 with latest recommended patches from 1 month ago.
>
> Also running Veritas VM 3.5 and Veritas FileSystem 3.5.
>
> Vmstat's "b" column, for kernel threads blocked for resources keeps,
> keeps climbing.
>
> Below is just one line of many:
>
> r b w swap free re mf pi po fr de sr s0 sd sd
> 1 38877704 0 4857296 1137056 262 1380 0 4 4 0 0 0 0 0 0
>
> As you can see, the 'b' column is currently 38+ million and climbing.
> What is more strange is that I would expect these numbers to manifest
> themselves as a system that would be completely unusable. The fact is
> that the system does not seem to be taking any kind of performance hit
> and is running along great.
>
> Obviously this number is incorrect but I can't find anything indicating
> why this number would be so high and growing.
>
> I have an identical system (V480) running the same version of Solaris
> and Veritas and with exactly the same patch levels which is not showing
> this strange behaviour.
>
> Strange?
>
> Any ideas?
>
> Thanks
>
-- John Malick Star Systems Engineering 232-236 West Market St. York, PA. 17401 john@starinc.com www.starinc.com ph:717-854-5911 fx:717-852-9421 _______________________________________________ sunmanagers mailing list sunmanagers@sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:30:30 EDT