2017-01-31 06:53 AM
Migrated off of a pair of mirrored EVA8000(P6000)'s to a snap-mirrored set of V3240s. (one with 600GB SAS, the other 2TB SATA) (NetApp Release 8.2.3P4 7-Mode) running over FC with Emulex LPe12000 HBA's.
VM Server access is really slow now, along with User data access.
VMWare ESXi Host (5.1U3) is showing a lot of latency errors like below.
Device naa.60a980004431396a375d444859356e31 performance has deteriorated. I/O latency
increased from average value of 4329 microseconds to 87418 microseconds.
1/31/2017 1:04:55 AM
Anyone have any ideas what might be going on, and what to look for to fix it?
2017-01-31 11:07 AM
I´d suggest to go for VMWare 5.5 latest U and ONTAP 8.2.4 latest P.
Then check the netapp interop matrix and the vmware matrix, usualy you need an updated fc driver for the hba which most often fixes any issues.
2017-01-31 11:25 AM
Thanks, but not an option. We're stuck with the HW we have for a while. I will check the matrix and the driver for the HBA, thanks.
2017-01-31 11:32 AM
Have you opened a support case to investigate configuration across your infrastructure?
At a minimum, check to see if there are any log messages on the ONTAP system which happen at the same time as the vSphere messages.
2017-01-31 12:17 PM
We're running NetApp Release 8.2.3P4 7-Mode, but when I go to the IMT, that doesn't show up, just P3. We've been running P4 for about a year now. Not sure why it wouldn't show up.
2017-02-01 06:27 AM
ESX 5.1 is EOS. That said, a perfstat and the VMware diagnostic bundle will be needed to find out the root cause. I would recommend opening a case with support after collecting the information.