08-29-2012 07:58 AM
Hello,
as mentioned in the title we are using Veritas Cluster 6.0 an therefore Veritas Volume Manger 6.0 on our Redhat 6.2 systems. A Clariion CX4-240 storage system provides some LUNs' for the servers. Everything is working fine apart from several I/O errros in the messages:
Aug 29 16:17:44 nwj8cnp001 kernel: end_request: I/O error, dev sdp, sector 0
Aug 29 16:17:44 nwj8cnp001 kernel: sd 8:0:5:0: [sdp] Device not ready
Aug 29 16:17:44 nwj8cnp001 kernel: sd 8:0:5:0: [sdp] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Aug 29 16:17:44 nwj8cnp001 kernel: sd 8:0:5:0: [sdp] Sense Key : Not Ready [current]
Aug 29 16:17:44 nwj8cnp001 kernel: sd 8:0:5:0: [sdp] Add. Sense: Logical unit not ready, manual intervention required
Aug 29 16:17:44 nwj8cnp001 kernel: sd 8:0:5:0: [sdp] CDB: Read(10): 28 00 3e 7f ff f8 00 00 08 00
All devices listed in the messages concern the secondary paths of the LUNs'.
We are using Failover Mode 1, does anyone know if this is causing those error messages? I have found the following information in an EMC document:
Failovermode may be set to 4 for ALUA behavior, or
Failovermode may be set to 1 for active/passive behavior
Do we have to change to Failovermode 4?
Regards,
Martin
Solved! Go to Solution.
09-30-2012 01:02 PM
VxVM supports both failover mode 1 and 4. Please see the Hardware technote page 10:
http://www.symantec.com/business/support/index?page=content&id=TECH47728
09-30-2012 01:02 PM
VxVM supports both failover mode 1 and 4. Please see the Hardware technote page 10:
http://www.symantec.com/business/support/index?page=content&id=TECH47728