[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

[Linux-cluster] GFS Problems



Hello all-

We're running RHEL 3WS and are up to date with RHN.  We run GFS
version 6.0.2.36-1 over iSCSI.

Lately we've had problems where the mounted GFS becomes unresponsive.
Commands like "df" never print the stats, nor return to the prompt.
They just increase the load on the machine.

Every time, there has been an error similar to the following:

Nov 30 04:28:42 buffalo kernel: lock_gulm: ERROR gulm_LT_recver err -110
Nov 30 04:28:42 buffalo kernel: lock_gulm: ERROR Issues sending state
request. -32
Nov 30 04:30:14 buffalo lock_gulmd_LTPX[2084]: ERROR [ltpx_io.c:1395]
XDR error -110:Unknown error 4294967186
Nov 30 04:30:14 buffalo lock_gulmd_LTPX[2084]: ERROR [ltpx_io.c:1379]
Client left before getting reply.
Nov 30 04:30:14 buffalo last message repeated 3071 times

The timestamp (approximately) coincides with the unresponsive filesystem.
When this error occurs, the mounted GFS on all 5 nodes become
unresponsive, not just the node where the error occurs.

Previously, we used GFS 6.0.2.27-0.1 and did not seem to have these
issues, though I am new to the organization and cannot say for sure.

Is this a known issue?  Does anyone know how I can resolve this?
Please help! :)
If you need more info, please let me know.

Thanks in advance!

-Joe


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]