VMware Cloud Community
IzzyR
Contributor
Contributor

VMs are removed from Resource Pools

I have recently been experiencing some odd behavior on one of my ESX servers. I have resource pools defined for different groups of servers under the host. The resource pools are not configured to use limits, memory/cpu reservations etc. As of now they are just being used as logical groupings. Every so often (second time this has happened in a month) some of the resource pools eject all VMs that are listed under them. I am unable to move the vm back into the resource pool. I do not receive any errors. Just a message in the tasks pane stating that the move into resource pool was completed. But the server never moves into the target resource pool. In order to move the vm back into the proper resource pools I need to first create another resource pool, then move the vm from the newly create resource pool into the original resource pool. The first time this happened I thought it was a fluke. But now that it has happened again I am not so sure. I was unable to find anything in tasks and events that indicates why the vm was removed in the first place. Has anyone seen this behavior before and know how to prevent it from reoccurring?

Reply
0 Kudos
3 Replies
Texiwill
Leadership
Leadership

Hello,

If you vMotion/migrate VMs from host to host you have to specify which resource pool to use during the migration. Did you migrate the VMs and forget this aspect? Was DRS used?

Did you collect a vmsupport file after the second time this happened? If not I would get one now, to preserve some state information. I would also open a case with your vmware support agent.

I have experienced a can not vMotion a VM into a resource pool before, when the resource pool was unlimited. But I could place it into the resource pool once the migration finished. Never a complete ejection of VMs.

Best regards,

Edward

--
Edward L. Haletky
vExpert XIV: 2009-2023,
VMTN Community Moderator
vSphere Upgrade Saga: https://www.astroarch.com/blogs
GitHub Repo: https://github.com/Texiwill
Reply
0 Kudos
IzzyR
Contributor
Contributor

I am not using vmotion (the servers have only local storage). And they are not being migrated, so there is no need to choose the resource pool. When the VMs were created, they were created in a specified resource pool. The VMs existed in the Resource Pools for months with no problems. Now all of a sudden, this problem arose.

Reply
0 Kudos
vhb-technician
Contributor
Contributor

IzzyR, the same happend here.

Using ESX 3.01 \[42829], local storage only.

We're new to the ESX environment so i can not give you any longtime experiences but it happend now the first time on both of our hosts.

It was not possible to move them again in the pool (altough the task was listed "in progress" and "completed"). On our live system only moving to a different pool first and then moving to the original pool was a success.

On our testserver we did a complete host reboot and then all machines outside the pools were automatically in the right pools again.

Could that be only a kind of display problem in the VC or VI-Client ?

I collected the vmsupport file, but don't know which of the log files could be interesting. Any suggestions where to search for ?

Reply
0 Kudos