Bug 175828

From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8) Gecko/20051111 Firefox/1.5

Description of problem:
We have 1 system with an Apple Xraid server attached via fibre channel connection. The system will write data fine locally without any problems to the array. We then exported the file system via NFS using tcp so our linux clients could write data to it. System runs fine for around 2-4 days without any problems. User submit large job that puts system under some higher load. Mounts will occur but then the system crashes without any errors or output. This is all we have in messages:

Dec 14 20:14:44 patemp1 rpc.mountd: authenticated unmount request from pace926:969 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:15:16 patemp1 rpc.mountd: authenticated unmount request from pace912:720 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:15:40 patemp1 rpc.mountd: authenticated unmount request from pace767:867 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:15:43 patemp1 rpc.mountd: authenticated unmount request from pace749:606 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:15:45 patemp1 rpc.mountd: authenticated mount request from pace737:645 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:15:52 patemp1 rpc.mountd: authenticated mount request from pace776:955 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:16:06 patemp1 rpc.mountd: authenticated mount request from pace930:879 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:16:10 patemp1 rpc.mountd: authenticated unmount request from pace755:600 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:16:35 patemp1 rpc.mountd: authenticated unmount request from pace938:805 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:16:41 patemp1 rpc.mountd: authenticated mount request from pace409:894 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:16:47 patemp1 rpc.mountd: authenticated unmount request from pace923:777 for /export/farmtmp1 (/export/farmtmp1)
Dec 14 20:20:41 patemp1 syslogd 1.4.1: restart.
Dec 14 20:20:41 patemp1 syslog: syslogd startup succeeded
Dec 14 20:20:41 patemp1 kernel: klogd 1.4.1, log source = /proc/kmsg started.
Dec 14 20:20:41 patemp1 kernel: ok
Dec 14 20:20:41 patemp1 kernel: Bootdata ok (command line is ro root=LABEL=/ console=tty0 console=ttyS0,9600)
Dec 14 20:20:41 patemp1 kernel: Linux version 2.4.21-37.ELsmp (bhcompile.redhat.com) (gcc version 3.2.3 20030502 (Red Hat Linux 3.2.3-53)) #1 SMP Wed Sep 7 13:32:18 EDT 2005
Dec 14 20:20:41 patemp1 kernel: BIOS-provided physical RAM map:
Dec 14 20:20:41 patemp1 kernel:  BIOS-e820: 0000000000000000 - 000000000009a000 (usable)
Dec 14 20:20:41 patemp1 kernel:  BIOS-e820: 000000000009a000 - 00000000000a0000 (reserved)
Dec 14 20:20:41 patemp1 kernel:  BIOS-e820: 00000000000d0000 - 0000000000100000 (reserved)
Dec 14 20:20:41 patemp1 kernel:  BIOS-e820: 0000000000100000 - 000000007ff70000 (usable)
Dec 14 20:20:41 patemp1 kernel:  BIOS-e820: 000000007ff70000 - 000000007ff77000 (ACPI data)

Would like more help or information on how to debug the system when the kernel crashes to determine what needs to be updated or changed. At this point we're running update 6 of WS 3 with the latest patches. If we can just get over this problem of unscheduled reboots/outages we would be running fine.

Version-Release number of selected component (if applicable):
nfs-utils-1.0.6-42EL

How reproducible:
Always

Steps to Reproduce:
1. Have 90 systems unmount and remount to 1 nfs linux server
2. mounts are using tcp/ip
3. The linux server eventually crashes under load. No error messages are displayed. Not sure on how to precede if I have nothing to debug, only see that the system rebooted.

Actual Results:  System restarts completely without any debugging information.

Expected Results:  System should just keep responding normally.

Additional info: