Hmm, what kind of behaviour that I am experiencing is that the server hangs and required restart to continue working. However, this behaviour occurs occasionally example after a few days or 1 or 2 weeks.
What I did was, checking the event logs to get a better understanding, what it shows on the event logs are “Event ID 27, 9 or 1076“, Source is “iScsiPrt”
The Message looks like this “Initiator could not find a match for the initiator task tag in the received PDU. Dump data contains the entire iSCSI header.” or “Target did not respond in time for a SCSI request. The CDB is given in the dump data.“.
With such little information was given from the event logs and I had a suspicious it could be configuration issue on the server. Research and investigate, but didn’t find any answers that cause this issue, so I had to visit the unit physically to see that there is an amber light on one of the disk but then is still not the conclusion to the messages from the event logs.
IBM Storwize has 2 controllers left and right, each controller has 2 management and iSCSI port. Each port supports as management and iSCSI port. Yea, I know it sounds weird, but it was true after I dug up their product guide. By default, it should be connecting to second port, first port is optional. However, for this case it was connected to first port.
I was trying out to access the management portal through Internet Browser, http based, but no valid. Possible suspicious that it could be someone touching the cable (cable changed), to answer that, is to try out each of the management port. Result, 4 of the ports on both controllers are not responding the management IP address that was configured for the IBM.
Well, that concludes that the management IP is no longer working, and please request IBM to reset the management IP and also order parts replacement for the amber light storage.
References:
Sabrina Kay's Blog 