We have a 4-server ESX 3.5 farm. I recently imported (P2V) a Windows server, and after cleaning it up (removing USB controller, Serial ports, etc...) I tried to power it up. I received the message "Could not power on VM; No space left on device. Failed to power on VM". I've checked all storage and have over 1GB free in most cases. Logs are only using a few KB. I can start and stop other VMs without a problem. This VM resides on a LUN with 2.4GB free, which is less than I'd like, but I can't see why that's not enough for minimal operation.
Here's where it gets weird. I move it to another host. Successful. I power it up. the VI client shows that it moves back to the original host, and then I get the same error message. Each time I try to move this to another host, it works until I power it up. Then it moves back! Same error again.
Outside of exorcising my hosts, what am I missing here?
It's exactly what I said to you ....
Don't forget to give some points if you're happy of my answer....
Regards
how much memory has the VM got? also are there any active snapshots on the other VMs on this LUN?
When you say 2,4 G free space ... You include swap file or not ?
If not, try to reduce memory to do a test ...
Regards
I think, this situation can occur when you have a lot of VMs running simultaneously and the disk space left is less. I mean when a number of VMs are running simultaneously, they will be using disk space for their respective swap files. In this situation powering on yet another VM can cause scarcity of disk space for swap files. so ESX is not powering on the VM - saying "no space left".
Yes, I verified it also. On my 10 GB volume I am having one 5GB linux VM. I filled rest of 4GB and 900MB space using "dd" command. Thus now 124MB was remaining on that voulme. My VM was having 256 MB memory configured in it. Now when I tried to power on the VM in VI client, it showed me the message "Insufficient Disk Space on Datastore". So if VM is not having much storage available to fulfill its memory/swap requirements, VI client would show the message.
If you found this or any other post helpful please consider the use of the Helpfull/Correct buttons to award points
Aha! I reduced the VM memory from 2.5GB (which is what the physical server had) down to 1GB, and it started immediately. Ok, so know I know the issue. I guess I need to read up on swap file space, because I'm unfamiliar with swap file resource requirements on ESX.
Thank you all for your help!
It's exactly what I said to you ....
Don't forget to give some points if you're happy of my answer....
Regards