DL380 G9 with P440ar Smart Storage Controller let Latency bounce in Esxi 6.5

[n4p]

Hi there,

i hope i found the right forum for my entry about the problem with the P440ar Smart Storage Controller.

If you take a look at the screenshot you will see that the latency from the configured Raid10 goes up and down.
It have nothing to do if there are any VMs up or all down.

After a little bit surfing through the web i found an info from HP which points on the nhpsa driver. -> https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-a00003625en_us

But with esxi 6.5U1 there is already a newer Version integrated. So what can be the problem for the latency going up and down?

Thanks for answer!
Kind regards

[stuart_xu]

hi,SIr

pls click the website to download the lastest the P440ar driver to fix the current issue.

https://support.hpe.com/hpsc/swd/public/detailsp4ts.oid=7274897&swItemId=MTX_0a27ab58458c46c8830a19c10c&swEnvOid=4184#tab-history

[n4p]

Thanks for your replay,

i think i found the fix for the issue. The P440AR is working correctly, the thing which let the latency go up and down is the HPE Dual SD-Card where ESXi is running on it.

I made a Test-Installation without the Dual SD-Card Stick and everything is fine. So what should i do now? Is that an defect from the stick or normal?

With an normal USB-Stick this latency up and down did not happen.

Thanks!

[stuart_xu]

hi,Sir

thanks for your reply,because we have your AHS log to check the root cause,so now we got the information that the OS was installed into the HPE 8Gb dual-SD card.

May I know did you check the vmkwarning.log,does the log will show the below error message

and occured the vmhba32 goes into an APD (AllPathDown) state intermittently. 

This issue ONLY occurs during runtime and will not occur during POST (Power On Self-Test).

The AllPathDown may occur intermittently if a routine is set up to access the microSDs or if the Operating System checks the availability of the microSDs.

The issue can be identified in the vmware support logs. The example below can be used to help and identify the signature of this issue:

First, a warning is received indicating that the vmkernel is experiencing communication issues with the USB device.

vmkwarning.log

2017-01-07T02:21:51.439Z cpu40:33490)ALERT: URB timed out - USB device may not respond

This is followed up by an indication of a path problem.

2017-01-07T02:23:08.580Z: [APDCorrelator] 4303195509883us: [esx.problem.storage.apd.start] Device or filesystem with identifier [mpx.vmhba32:C0:T0:L1] has entered the All Paths Down state.

if yes, may be you should check the SD-card  firmware in the ilo UI. then upgrade the firmware to lastest version.

or you should call the local HPE support center to replace the HPE SD-card.

https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_c8d0b3a40eed48d4a622d0dfd7

[n4p]

Hi,

thanks for your answer, but i didn't found any log entries like this.

Is there any possibility to find out what version i have on the SD-Card Reader?

Thanks