Hi, I have a Esxi 6 with some VMs, local datastore (ssd drives). For one of my VM, in vsphere from vcenter server, I have the standard warning saying that I need to do disk consolidation for that server. I have checked in snapshot manager and no snapshots. When I try to consolidate, it gives me an error saying (it is in french I will translate the best I can);
Impossible to send a bitmpa (this is not a typo, it's not bitmap, it's really bitmpa) allocated to vmkernel while copying the disk xxxxxxxx.vmdk to xxxxxxxx.vmdk
Did some research and did not really find anything exactly with this error but I have verified that no backups are being made (we use veeam), and tried again after I have stopped all veeam services on the vcenter server. I have 3 virtual disk on that server if that matters. My next guess would be to stop the virtual server and try the disk consolidation again but I have to wait this weekend to do this and I don't know why I have the feeling that I will get the same error.
The VM works fine, I can do backups but I have a warning in the veeam backup notification for this server;
"server has stuck VM snapshot, will attempt to consolidate periodically"
If anyone has any idea on what I should try it will be appreciated.
A few things can cause this. To solve try the following if possible:
Let us know the outcome, I hope this helps.
Hi, unfortunately nothing worked. I can't do a Vmotion since I don't have enough space (the server has 3 partitions that totals about 3-4 Tb. Now though I have a new error, which is "an error occurred while consolidating disks 5 (input/output error)". This happens when ALL vm's are down and only Esxi up, connected vsphere from my workstation. As soon as I boot all VM and try again from vcenter, I have the original error message that I posted at the beginning.
If I browse the datastore, I see normal content for virtual 😧 and E: drive, but for the C: drive, I see a lot of xxxx.vmdk and ....ctk.vmdk (18 actually) which looks to me like snapshot but there is nothing in the snapshot manager. By the way I can create a snapshot and delete it no problem.
I just got the ok from the client to call paid vmware support so I will call them probably tomorrow. Until then if any other ideas go ahead.
So the problem is definitely that you have snapshot files that have not been cleanly consolidated.
Please post the VM's vmware.log files and also the hostd log file on the ESXi host.
Between them we should find out what the problem is and find a resolution.
Do you see in the error anything related to change tracking?
I asked for them all as some might contain errors and others not. It's always best to check them all to be thorough.
Hi, no the unmount of the cdrom changed nothing. I contacted the vmware support, the first thing they ask me to do is to retry the consolidation when the VM is powered otf, but they asked me to rename the .vmsd file before the consolidation. Since I don't have a good time frame during the week to do this, the client asked me to try it next weekend.
My gut feeling tells me it will not work but I guess it's worth a try. I'll post next week.
I think you said you use Veeam? I have had veeam cause this scenario. Look to see if there is a drive mounted to your veeam guest / server via edit settings on the veeam guest itself. If so note the path of the hard disk, If so, you would need to get rid of that disk to clear the consolidate error.
Hi, unfortunately what the vmware support suggested did not work. In an email, he was asking me to get the support-vm logs. Having no explanation on how to do this, I googled it and after went in vsphere, file, export, system something...
The problem is that the only thing that was there is the option "hungvm". So I selected it. What happened is that it crashed the VM entirely, so we used the replicated version of the VM on another Esxi. We lost one day of data, but at least I don't have the snapshot/consolidation on this one. So I guess case closed.
Thanks all for your help.