I have a reproducible hang with kswapd in the run queue, everything else is
in an i/o wait. The load average is climbing.

Using either a highpoint RR2340 or an LSI8888ELP PCIe-8 lane controller,
I max out the write rate at between 900MB/Sec to 1.1GB/Sec into
16 seagate 500GB ES series drives. Eventually the system locks up
with kswapd0 getting 100% of one CPU. kswapd1 is not running.

The system is a SuperMicro H8DM3-2 (or an H8DMi-2) with 2222SE or 2216 
opterons,
16GB of RAM. 2.6.22-5, 2.6.21 and 2.6.23-rc7 and 2.6.23-rc9 all lock up.
2.6.20 does not, but it also runs 200MB/Sec slower in write rates.
The base O/S is Centos 5.0
I can patch in KDB and look around (did this for 2.6.22-5) but I'm
not sure what to look for in kswapd to see what got lost to keep
the system locked up. With eralier kernels, the system needs a reset
button to recover. With 2.6.23-rc9 I was left with enough to get the 
following
ps, top, and /proc/meminfo data

Hints anyone (please) as to how to slay this dragon?

Berkley


-- 

// E. F. Berkley Shands, MSc//

** Exegy Inc.**

349 Marshall Road, Suite 100

St. Louis , MO  63119

Direct:  (314) 218-3600 X450

Cell:  (314) 303-2546

Office:  (314) 218-3600

Fax:  (314) 218-3601

 

The Usual Disclaimer follows...
 
This e-mail and any documents accompanying it may contain legally privileged and/or confidential information belonging to Exegy, Inc. Such information may be protected from disclosure by law. The information is intended for use by only the addressee. If you are not the intended recipient, you are hereby notified that any disclosure or use of the information is strictly prohibited. If you have received this e-mail in error, please immediately contact the sender by e-mail or phone regarding instructions for return or destruction and do not use or disclose the content to others.