Guest CPU %RDY >30 & %WAIT > 1000

[mkielman]

All -

I have a guest with a %RDY >30 and a %MLMTD = 0. It is also maintains a %WAIT of about 1000 (%IDLE is approx. 270). Using the thresholds listed at this site: http://www.yellow-bricks.com/2010/01/05/esxtop-valuesthresholds/ I am not seeing any Disk latency to worry about all the "MCTL" counters are 0 except MCTLMAX which I understand is nothing to worry about. The problem is, the end users are complaining of periodic performance issues and the OS counters are not pointing to anything unusual EXCEPT CPU is roughly around 30% with a Proc Queue of about 5 per processor. This guest is configured with 4 procs.

Any assistance would be greatly appreciated.

Thanks!

[mkielman]

Ok, I just moved the guest to another host in the same cluster and now it isn't even showing up in ESXTOP so I can't see what the %RDY is. Is there any other way to grab this info?

[danm66]

are you sure DRS didn't move it to another host? does vm-support -x list the vm?

[mkielman]

Yes I am sure.

[mkielman]

I wanted to revisit this question because I have another guest with similar symptoms, however, it is our patching server so I can reproduce the "performance" issues.

Here are the details:

• Application confirmed to be multi-threaded

• Originally running with 2 vCPUs and the Proc Time (according to OS) was 100% and Processor Queue Length of approx. 35 when we were running the workload

• Increased to 4 vCPU, Proc Time (according to OS) reduced by 30%, however, queue length remained the same and the "sluggish" performance perisisted. At this time, I checked esxtop. When it was running on 1 of the hosts, the %RDY peaked at 175%. While I was looking at it, DRS moved it to another host and since then, it hasn't showed up in esxtop with regard to any metric (network, memory, cpu, disk)

I guess my point here is, tracking performance problems is fairly difficult and %RDY although has been described as something which needs to be below 30, doesn't seem to indicate a problem when the slowness occurs whether or not %RDY is high.

Are there other suggestions? What concerns me the most is the Queue lenght as reported by the OS.

Please help!

[unnikpr]

HI,

Is this resolved. I am facing something simile with Vm running SAS. The bare metal installation gives nice performance but same configuration VM have seconds difference performance issues. This is a Linux 7. and CPU is the bottleneck. the I/O are all good. But SAS operation simply slow down, when i decrease the CPU from 32 to 8 it's getting better.

Regards,

Unni