Hello,(~1000 lines snipped)
I apologize for the long post. I tried to include as much
relevant information as possible.
Thank you,
Eric
[1.] One line summary of the problem:
machine oops under heavy I/O or network access ???
[2.] Full description of the problem/report:
I have a Supermicro X6DHE-XG2 (Intel E7520 chipset) with dual Xeon with 4GB RAM.
I am using one 3ware 8005 controller, three 3ware 9550sx controllers, and
reiserfs. This machine is acting primarily to serve NFS.
The oops below are from my original machine configuration, but I have tried other some variations and have other oops:
- I have tried Suse 10.1 kernels 2.6.16.13-4-smp and 2.6.16.13-4-default.
- I have rebuilt the system onto an identical spare hardware configuration.
- I have used 64-bit and 32-bit kernel versions.
- I have replaced all the shared reiserfs filesytems with ext3.
In all cases under load, disk IO alone (NFS server off) or disk IO+NFS operations, the system has oops and "general protection fault" errors.
The process is usually nfds or rpc.mountd, but (with the NFS server turned off) I
also have errors from, for example, cp and rsync. The system has been up for as much as 16 hours with no problems, and sometimes as little as five minutes.