Veeam Job Powered Off Server!

[klasikrkade]

I have a replication job that runs on this particular server once a day 7 days a week.

Over the weekend it failed 4 times and I did not have time to remediate it until this morning. I noticed that the server was powered off.

Looking in the datastore for this box I noticed that it had tons of snapshot files. These did not go away after deleting the consolidated snap and the normal veeam snap in snapshot manager.

I tried creating another snapshot which worked and then deleting it thinking that this would perhaps clean up all of these extra snaps. No dice.

I tried cloning the box and this failed.

I ended up deleting the 2 disks from withing the servers properties in vcenter. I then added the 2 original disks, choosing them from the datastore that the server was on. I then deleted all of the extra snaps and powered on the box successfully.

This is the second server here that powered itself off due somehow to veeam snaps.

Can someone give me some insight as to why this is happening?

[jpaul]

VMware only supports a certain number of redo logs (snapshot files) After you get more then that number (32) then vmware powers off the virtual machine and refuses to turn it back on until you fix the problem.

You have a couple options from that i recall, one of which was to use ESX tools to clone the drive to a new VMDK file which will basically consolidate all the snapshots into a single vmdk.

I would give VMware a call on this one though, cause if not done properly your data will go bye bye

[Vitaliy S.]

Furthermore, I would recommend looking through VMware's VM log files (vmware*.log) at the time when VM powered off. Should be helpful.