I just started my second week as a sys admin at a new job and have been cleaning up snapshots the previous admin left behind. I’m now down to the last two servers which is giving me pause.
One is the domain controller, and AD is my sole means of access to vSphere. There is no other domain controller. There are three snapshots on this server from PowerCLI:
Name : 12142020newserver |
Description : |
Size : 0.01 GB |
Created : 12/14/2020 5:54:56 PM |
Name : SERVER122021 |
Description : |
Size : 48.01 GB |
Created : 1/2/2021 3:32:54 PM |
Name : Working_5_7_2021_12_44_PM |
Description : |
Size : 1,178.15 GB |
Created : 5/7/2021 11:43:30 AM |
The domain controller has two thick provisioned disks (below in bold). One is 2TB and the other is 400GB. The total approximation of disk usage for this VM and its snapshot is 3.63TB
For reasons listed below, here's the disk usage I captured in the datastore:
GB | |
DC_1.vmdk | 2048.0 |
DC_1-000002.vmdk | 883.0 |
DC.vmdk | 399.9 |
DC-000002.vmdk | 245.8 |
DC-000006.vmdk | 49.2 |
DC-000001.vmdk | 47.9 |
DC-Snapshot.vmem | 15.4 |
DC-af54141.vmsp | 15.4 |
DC-Snapshot1.vmem | 15.4 |
DC-1-000006.vmdk | 0.2 |
DC-1-000001.vmdk | 0.1 |
The other is an application server. There was no intentional snapshot but Veeam 10 created the “Veeam Backup Forced Consolidation Snapshot” on 9/18/2021. PowerCLI shows the snapshot is currently 984.02 GB.
Description : Veeam Backup Forced Consolidation Snapshot |
Size : 984.02 GB |
Created : 9/18/2021 7:45:24 PM |
This server has thick provisioned disks as well (below in bold). One with 1000GB and the other with 250GB. The total approximation of disk usage for this VM and its snapshot is 2.2TB
Here's the disk usage I captured:
GB | |
SERVER_1.vmdk | 1000.0 |
SERVER_1-000003.vmdk | 934.8 |
SERVER.vmdk | 250.0 |
SERVER-000001.vmdk | 49.3 |
SERVER-d3277b5e.vswp | 16.0 |
vmx-SERVER-354580062-1.vswp | 0.2 |
vmx-SERVER-354580062-2.vswp | 0.2 |
The datastore shows 7.63TB Used and 1.46TB Free. There are two additional servers on this datastore, server2 using 1.15TB and server3 using 632GB.
***I do have two other hosts, each with its own datastore. One with 2.71TB free and the other with 3.15TB free. I don’t know what the best practice is here if I want to move server2 and server3 to a different datastore and/or host. All three hosts are Dell servers with their own local storage as their respective datastore.
Additional info:
vCenter 6.0.0, 3634794
ESXi, 6.0.0, 36270759
I do not have root access. I’m accessing vSphere with AD creds.
I cannot SSH into vCenter or the host. I enabled SSH on the host but cannot escape the backslash properly in Windows and with the iOS Terminus app, it appears that the NetBIOS user is correct but I cannot authenticate with my AD creds.
Thus, I cannot provide the “ls -lisah” output.
I cannot vMotion running servers because I do not have the licensing to do so. I cannot activate the evaluation license.
What do I do first?
I am not a super vmware pro user, but have been using vsphere essentials since 4, and am just moving to esx 7.
First off, always have 2 AD servers just in case one is rebooting or your esx goes down that is running AD. Redundancy is your friend as a sys admin, even for your internet connection (why all the ISP will be calling you to sell you SDWan)
You can use veeam to move the other VMs to a different esx host using live migration. It sounds like you have an essentials license like me, but veeam can do some of the cooler stuff like host and datastore migrations. Make sure you have a current veeam backup of your VMs.
With the VMs themselves being so large, I would leave them be and move the smaller vm to free up space if your worried. Then clean up the vm that uses less disk space first just in case anything expands or your worried.
removing snapshots shouldn't use more disk space, it should free it up. I was just in your same boat with VSS bugs making me think my exchange server was going to run out of space and planning on moving VMs around ( 2tb iSCSI LUN that I made a long time ago looked 90% full from the server storage growth (thin disk, but they expanded). fixed the VSS bug, now I have 90% open drives from 90% open datastore).
they probably have a putty SSH client for mac for your ssh issue (escaping something).