VM "Uptime" resets after migration

[m_d_sella]

I am currently running ESX 3.0.2 with Virtual Center 2.0.2. While testing DRS, I have noticed some unusual behavior in Virtual Center. When I place a host into Maintenance Mode, DRS handles all of the VM migrations as it should and distributes the VMs across the remaining hosts in the cluster. When these migrations occur however, Virtual Center resets the "uptime" of each VM to zero as it migrates. It seems to me that a migration should not affect the uptime statistic for a VM in this fashion. As such, it is not a critical issue, but ultimately I would like to provide reports to our executives that adequately represent the VM availability that we are seeing in our environment.

Has anyone else seen this issue?

Thanks,

Mike

[esiebert7625]

It's a feature, uptime is tied to the host it is running on and not the OS of the VM.

http://www.vmware.com/community/thread.jspa?messageID=644216򝑸

Fyi…if you find this post helpful, please award points using the Helpful/Correct buttons.

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Thanks, Eric

Visit my website: http://vmware-land.com

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

[m_d_sella]

esiebert,

Thank you for the reply and link.

Do you happen to know if this was the case in ESX 2.5.x as well? I no longer have a 2.5.x host to test with, but I am pretty confident that at some point the counters did not reset after VMotion (possibly prior to my upgrade to VI3). Just curious, as it seems kind of odd to reset those counters in this case, especially seeing that the hosts have their own uptime counters. A reset of the uptime counter would imply that the VM was not available, and I'm not sure what value "uptime since last VMotion" would provide. I do not doubt that this is the case, but am simply curious if I am missing something behind their reasoning.

Thanks,

Mike

[esiebert7625]

Not sure on that, DRS is new to ESX3 as is responsible for alot of VM moves by Vmotion compared to ESX2. Previously in ESX2 VM's typically only moved in case of host failure or maintenance. So it may have been changed because of DRS.

[m_d_sella]

Opened a call with VMware support on this topic to get their opinion and/or open a feature request.

Turns out their are two active bug reports that have been submitted to VMware Engineering on this issue. Ultimately, the cause appears to be a limitation relating to the transfer of the uptime stats from one host to another, due to the VM changing "worlds" during the migration.

Thanks for the assist, esiebert.

[esiebert7625]

No problem...