Target path has gone offline/online problems on 11.31

[Telia BackOffice]

Hi.

I have upgraded a bunch of our machines to 11.31. They work nicely, but I get errors about target path online/offline that I did not have before the upgrade. Can you help me figure out why?

Jul 22 15:30:20 riker vmunix: Target path (class=tgtpath, instance=3) has gone offline. The target path h/w path is 0/4/2/0.0x5006048c49ae4127
Jul 22 15:30:20 riker vmunix: class : lunpath, instance 15
Jul 22 15:30:20 riker vmunix: class : tgtpath, instance 3
Jul 22 15:30:20 riker vmunix: An attempt to probe existing LUN id 0x0 failed with errno of 22.
Jul 22 15:30:20 riker vmunix: class : lunpath, instance 31
Jul 22 15:30:20 riker vmunix: Evpd inquiry page 83h/80h failed or the current page 83h/80h data do not match the previous known page 83h/80h data on LUN id 0x0 probed beneath the target path
(class = tgtpath, instance = 12) The lun path is (class = lunpath, instance 31).Run 'scsimgr replace_wwid' command to validate the change
Jul 22 15:30:20 riker vmunix: An attempt to probe existing LUN id 0x0 failed with errno of 14.
Jul 22 15:30:24 riker vmunix: class : lunpath, instance 3
Jul 22 15:30:20 riker vmunix: class : lunpath, instance 31
Jul 22 15:30:24 riker vmunix: An attempt to probe existing LUN id 0x0 failed with errno of 22.
Jul 22 15:30:24 riker vmunix: class : tgtpath, instance 3
Jul 22 15:30:24 riker vmunix: Target path (class=tgtpath, instance=3) has gone online. The target path h/w path is 0/4/2/0.0x5006048c49ae4127

[Michael Leu]

Hi,

I think the important part is this:
Jul 22 15:30:20 riker vmunix: Evpd inquiry page 83h/80h failed or the current page 83h/80h data do not match the previous known page 83h/80h data on LUN id 0x0 probed beneath the target path
(class = tgtpath, instance = 12) The lun path is (class = lunpath, instance 31).Run 'scsimgr replace_wwid' command to validate the change

Which AFAIK means the WWID of the LUN is not the same as the cached WWID.

# scsimgr replace_wwid -C lunpath -I 31
This should update the lunpath with the new WWID.

To see the current WWID you can use
# scsimgr get_attr -H -a wwid

[Steven E. Protter]

Shalom,

Looks like PVLINKS is active. There are two paths to the LUN and your system is switching off.

Or the single path to the LUN is going off line.

Possible causes:
1) Bad HBA card
2) Bad fiber cable
3) Problem on the fiber switch.

All need to be investigated.

SEP

[VINY]

Was there ever a resolution to this? I am having the same problem and not sure if it is the cable. We ended up direct connecting the server to the array and still have the same problems.