HP StorageWorks Scalable File Share - Manual (Page 265)

Troubleshooting file system performance 9–41
Remedial action
See Section 9.29 for information on troubleshooting LUN presentation.
9.27.5 Examining the system logs for errors
Events that occur on any server in the system are sent to the administration server (or to the MDS server if
the administration server has failed) and logged in the system event log. You can use the
show log
command and an appropriate filter to look at the logs for each server in turn, as shown in the following
example:
sfs> show log server=south2
where
south2
is the name of the server whose events you want to examine. (For more information on the
show log
command, see Section 4.6.1 and Section A.20.9.)
Run the command for each server. In the output from the command, look for any Lustre errors or SCSI errors
such as those shown in the following examples:
2004/10/18 10:25:01 kern ! south2: LustreError:
1983:0:(ldlm_lib.c:1241:target_send_reply_msg()) @@@ processing error (-16)
req@d3354800 x9/t0 o38->@:-1 lens 168/64 ref 0 fl Interpret:/0/50000 rc -16/0
2004/10/15 08:01:01 kern ! south2: LustreError:
466:0:(qswnal_cb.c:485:kqswnal_txhandler()) Tx completion to 0x3 failed: -3
2004/10/12 14:05:01 kern ! south2: LustreError:
18316:0:(ldlm_lib.c:1241:target_send_reply_msg()) @@@ processing error (-107)
req@f57df000 x12646679/t0 o400->@:-1 lens 64/0 ref 0 fl Interpret:/0/0 rc
-107/0
2004/10/21 07:15:01 kern ! south1: LustreError:
18317:0:(ldlm_lib.c:1241:target_send_reply_msg()) @@@ processing error (-19)
req@f57df800 x12646680/t0 o38->@:-1 lens 168/0 ref 0 fl Interpret:/0/0 rc
-19/0
2004/10/15 08:25:01 kern ! south10: LustreError:
3557:0:(client.c:203:ptlrpc_prep_req()) ASSERTION((unsigned long)imp > 0x1000)
failed
2004/10/15 08:25:01 kern ! south10: LustreError:
3557:0:(module.c:71:kportal_assertion_failed()) LBUG
The following examples are messages that indicate events such as the file system being started or stopped,
and services failing over or failing back. Such messages are not necessarily indicative of a problem,
although performance will be degraded if the services continue to run in a failed-over configuration:
2004/09/15 16:25:02 kern ! south2:
LustreError: 15875:0:(mds_fs.c:521:mds_fs_cleanup()) down-mds3:
shutting down for failover; client state will be preserved.
2004/10/17 20:45:51 kern ! south3:
LustreError: 15875:0:(mds_fs.c:521:ds_fs_cleanup()) south-mds3:
shutting down for failover; client state will be preserved.
2004/10/22 22:15:43 kern ! south3:
LustreError: 4003:0:(filter.c:1372:filter_cleanup()) south-ost8:
shutting down for failover; client state will be preserved.
Remedial action
Address any errors being shown in the client or server logs. Use the information provided in Chapter 4 and
in other sections of this chapter (Chapter 9) to assist you in understanding and addressing the errors.