I have a customer testing out a workflow that lets them control a single use snapmirror and this week it seems to have failed in a spot that I cannot find a reason for.
I am running WFA 126.96.36.199.32.
We Set Volume State, Create VSM, sleep 60 seconds due to a known issue with 2.1, Wait for VSM, Break VSM, then do custom changes to the volume.
During the run this past week, the Wait for VSM appears to have reported complete without being complete and Break VSM ran and the job failed. After that the job could not be resumed because it kept reporting there was no snapmirror relationship.
Wait for VSM reported the following KB transferred, but not complete, then break VSM ran: 172039040 KB
Snapmirror log reports the following KB once complete 17 mins later: End (189780128 KB)
Any ideas on what caused this would be greatly appreciated as we are replacing a homebrew script that requires three teams with WFA with this.
Since this would be a new snap mirror relationship you need to use the finder for a "new snapmirror relationship" vs a existing relationship. WFA pulls that info from the reservations tables vs the normal locations.
The VSM relationship is defined in step 2, that definition is then used in step 4 so there is no need search for the VSM. This job is one of 5 built like this and it is the only one that has ever failed in this fashion.
The controller logs and wfa log clearly indicate that Wait for VSM was not complete(the last Wait for VSM update in the WFA log is uninitialized, transferring), but WFA indicated that it was and it could proceed to Break VSM which was not correct.