04-03-2024 04:26 PM
Hi all,
As vMotion is officially supported on the C9800-CL as from release 17.9.2, we have tested it on a few HA deployments and it works well, however Cisco docos state: "Do not run vMotion on both the active and standby VMs at the same time. In the time it takes for the active and standby to move to the new hardware resource, the 9800-CL may be seen as going down".
This makes perfect sense, however in one of our deployments I was told this is not possible:
"Both 9800-CL VMs are in the same vCenter, but on completely separate host clusters, at physically separate data centres.
We can only configure rules affecting VMs at the individual cluster level, not higher. We can put an override in place to disable full DRS automation, but that wouldn’t be recommended since that could cause other problems if a host fails for any reason."
Has anyone faced this challenge before, and how was that solved?
Many thanks!
Feds
04-03-2024 10:26 PM
- Since this is a policy problem , I can't see how it could be solved at the (technical) C9800-CL level ,
M.
04-03-2024 10:35 PM
Hi Marce1000,
Thank you, and I agree, this question would be more relevant in a VMware forum.
Nevertheless someone may have faced this for 9800-CL and found a fix or a workaround already, I'm just checking..
And if I ever find a useful solution to this problem I'll post it here.
Thanks
Feds
04-03-2024 11:00 PM
- The way I see it , if there would be no way out and there is an important demand on this from business ; is to then 'fallback' to physical (9800) boxes ,
M.
04-03-2024 11:07 PM - edited 04-04-2024 01:27 PM
Definitely.
Another option would be potentially and temporarily sacrifice redundancy by disabling hot vMotion for one of the VMs, and do that manually if the need arises.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide