Views:

Summary



An incomplete iSCSI session leads to a job failure in a Linux environment.

Symptoms



From the job log:

xxx.xxx.xxx.xxx 1/25/2016 8:35:13 pm SNBAPP_004X java.lang.Exception: getDRSystemInformationN: GetSystemInformation failed, rc (-1), reason (ERROR: Error calling get_partition_info() for SCSI Disk ...)

xxx.xxx.xxx.xxx 1/25/2016 8:35:13 pm SNBAPH_126E Func(AHOpPfs:run): Exception (com.syncsort.bex.msg.bexException: 21005 )

From /var/log/messages of the IA-map destination:

Jan 25 21:19:21 myhost LINUX-DRAPI[24729]: ERROR: MBR Read Device[/dev/sdb] failed

Jan 25 21:19:21 myhost kernel: Buffer I/O error on device sdb, logical block 0

Jan 25 21:19:21 myhost kernel: Buffer I/O error on device sdb, logical block 1

Jan 25 21:19:21 myhost kernel: Buffer I/O error on device sdb, logical block 2

Jan 25 21:19:21 myhost kernel: Buffer I/O error on device sdb, logical block 3

Jan 25 21:19:21 myhost kernel: sd 28:0:0:0: SCSI error: return code = 0x08000002

 


Resolution



  1. Verify if there are valid iSCSI sessions in a Linux iSCSI destination node. Run iscsiadm -m session -P 3.
  2. If iSCSI sessions are not valid, or iSCSI sessions can be removed, run iscsiadm -m node -T [iSCSI target hosting the Lun] –p [IP address of iSCSI target]:3260 --logout
  3. If iSCSI sessions include any valid sessions, consult with the owner of the valid session and schedule a maintenance time for removing the session.