Recently I had the "funny" task of moving a lot of volumes from some old aggregates to new ones... I noticed that most of the volumes star out "hot" but towards the end it just sits there with like 50MB left, and a speed of 5-8MB/sec.. and just staus there for like 30 mins... I cannot see any issues... it just takes forever to complete... I changed to "govoner" to just allow 4 jobs which is two moves in this setup because we move to/from the same controller... But it's just frustrating to see it sitting there... eventually it completes of cause... but still... is there something one can do in order to speed this up? This is a cluster with mostly snapmirror destinations, so not real load on the system, other than this volume move thing... suggestions or an explanation would be nice 🙂
#Why do some volume moves take so long to complete?
jagged cypress
fossil grove
Finalizing phase might take longer. And cutover might need to retry if it's a busy volume.
Also when you have a high number of snapshots it takes longer.
Are you changing the tiering-policy with your vol move?
jagged cypress
Well there are neither... the volumes aren't busy and I cannot see any retries happening... also I cannot see any snapshots beeing updated.. so I don't think this is it... We are moving from a FlexCache aggregate to a normal aggregate (but with aggregate encrytion enabled)...