Lets say we have 20 linked clones in pool a that has refresh immediately after logoff. 10 people are logged on to VM's 1-10. When we schedule a recompose of the entire pool, we would expect that VM's 11-20 would recompose after which, as people log off of Vm's 1-10, those VM's would recompose to the new replica. Instead, VM's 1-10 are refreshing to a checkpoint and never recomposing to the new replica. It seems that when refresh after logoff was not set, the recompose was working correctly. Are these two features mutually exclusive?
I'm assuming this is what you are talking about? http://kb.vmware.com/kb/2006707
If so, yes, this is a known issue, a refresh on log off will supersede a recompose on log off. I've escalated this issue until I was blue and have heard it will be fixed in a future version
Right now I'm manually doing recomposes by going into Pool > Inventory Tab > Desktops (View Composer Details) > Right click desktops > Recompose. I usually do this on a single desktop and let the replica create, and the do batches of recomposes afterwards. Also recomposing and selecting force users to log off will do the trick too.
Moderator note: Edited the post to remove NDA restricted information.
I'm assuming this is what you are talking about? http://kb.vmware.com/kb/2006707
If so, yes, this is a known issue, a refresh on log off will supersede a recompose on log off. I've escalated this issue until I was blue and have heard it will be fixed in a future version
Right now I'm manually doing recomposes by going into Pool > Inventory Tab > Desktops (View Composer Details) > Right click desktops > Recompose. I usually do this on a single desktop and let the replica create, and the do batches of recomposes afterwards. Also recomposing and selecting force users to log off will do the trick too.
Moderator note: Edited the post to remove NDA restricted information.