It seems as if this was a Kernel problem plaguing 2.6.9-55.0.9 (RH)
We upgraded to 2.6.9-67.0.4 and the problem went away.
-- Christer
On 21 feb 2008, at 14:25, Christer Bernérus wrote:
Hi folks.
Anyone running RHAS4 systems who are seeing processes hung in
block_congestion_wait?
This shows in top like:
top - 14:20:53 up 22 days, 2:28, 51 users, load average: 5.83,
5.75, 6.69
Tasks: 717 total, 1 running, 705 sleeping, 1 stopped, 10 zombie
Cpu(s): 0.2% us, 0.3% sy, 0.0% ni, 49.6% id, 49.9% wa, 0.0% hi,
0.0% si
Mem: 16573732k total, 11114520k used, 5459212k free, 5620k
buffers
Swap: 18872224k total, 35340k used, 18836884k free, 8388056k
cached
(See the "wa" value on the third line) On our systems, this
eventually reaches over 99%, where very little useful work is done.
Running ps laxwww shows many processes hanging in on WCHAN blk_co.
These processes are completely stuck for a long time.
Eventually, the situation clears by itself. Priobably because users
give up.
Christer Bernérus
Chalmers University of Technology, Sweden
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
