I'm running WFA 126.96.36.199.0-2879153 with cDOT 8.3.1RC1 (vsim).
I run a break workflow on an intercluster SnapMirror relationship. After the break is complete, I run a reverse resynch which finishes successfully. Later, I break the new reversed relationship and run another reverse resync to hopefully return the relationship to its original state. The second reverse resync fails on the state check (claims state is not "broken-off"). From everything I can see via OCUM, SM, and clustershell, the relationship is indeed broken. If I edit the workflow to skip the state check, it runs fine.
I'm initiating all of this via OCUM, but doing it from WFA portal makes no difference.
Has anybody seen this, or have any idea on what might be happening?
I would be willing to upgrade to a newer build of WFA if I think it will solve the problem, but I depend on others for this environment so trial and error might not be the best approach.
I did manage to slip in an upgrade to WFA 3.1P1. Now it's getting the same failure on the first run on reverse rsync--it's not waiting til the second.
Here's the description of the failure:
Track execution progress of workflow 'Reverse Resync SnapMirror'. Failure Reason:
[WFA reported] Workflow aborted. SnapMirror relationship to SVM_black:/SVM_white_Admin_2017_SnapMirror_001 must be in a broken-off state. Break the SnapMirror relationship and then rediscover the destination cluster in OnCommand Unified Manager. Location: 'Search for relationship' command > 'SnapMirror' tab > 'forward_snapmirror_relationship' variable
Here's the state of the relationship before running the reverse resync: