How to report an issue with custom HPE VMware image?

[dtango]

Hello.

Is it possible to report an issue with custom hpe VMware image (6.7) to HP?

Example: I've got a brand new DL380 Gen10 with SR816i.

I've created 2 volumes (the same number of hard drives in RAID, the same size), and when I checked how Queue Depth settings look with standard and custom HPE Image I see the difference.

This doesn't look very critical,  but still, settings in HPE  image doesn't look right to me.

Any option to report this to HP? 

KR,

Aleks.

[dtango]

Ok, If I cant report to HP, is it possible to change  "Queue Depth" parameter with a CLI command? 

Should I open a support case with VMware?

KR,

Aleks.

[RaviTyagi]

Hello Aleks,

Greetings !

Please find below the information as requested .

Queue depth is the number of commands that the HBA can send / receive in a single chunk - per LUN. When you exceed the queue depth on target, you should expect performance degradation. When too many concurrent I/Os are sent to a storage device, the device responds with an I/O failure message of queue full (qfull). This message is intended to cause the host to try the I/O again a short time later

In general, the number of servers an array can adequately support is dependent upon following things:

• Queues Depth per Physical Storage Port.
• Number of Storage Ports.
• Bandwidth (throughput) per Controller.

(Queues Depth per Physical Storage Port x Storage Ports) / Host Queue Depth = Number of servers

So for example, we have an 8 port array (4 ports per controller) with 1024 queues per storage port and a host (LUN) queue depth setting of 64 will be able to connect up to:

( 8 x 1024) / 64 = 128 single connected servers or 64 (2 x HBA) or 32 (4 x HBA)

The fan-out ratio is 16:1 meaning 16 initiators per storage port.

What's about VMware infrastructure and multiple ESXi hosts communicating with the storage ports? The Queue Depth should be calculated by the following formula:

Port-QD => ESXi Host 1 (P * QD * L) + ESXi Host 2 (P * QD * L) ..... + ESXi Host n (P * QD * L)

where:

Port-QD  = Target Port Queue Depth
P = Number of host paths connected to the array target port
L = Number of LUNs presented to the host via the array target port
QD = LUN queue depth on the host

As shown above, it is recommended that the total aggregate of the maximum number of outstanding SCSI commands from all ESXi hosts connected to a port of an array should be less than the maximum queue depth of the port.

Please find below the relevant KB article .

Changing the queue depth for QLogic, Emulex, and Brocade HBAs =>(https://kb.vmware.com/s/article/1267)

You can reach out to Support using the following URL :https://www.hpe.com/us/en/contact-hpe.html

Regards,

RRT

I am an HPE Employee 

Disclaimer: One or more of the links above may take you outside the Hewlett-Packard Enterprise Web site, HPE does not control and is not responsible for information outside of the HPE Web site. Third-party links are provided to help you find the support you need. This information is subject to change without notice. We would be unable to assure the accuracy of this third-party information. The software manufacturer will be able to provide support for this third-party software.

Data integrity is customer responsibility. HPE recommends to have data backup frequently and before performing any activity so that it can be restored during disaster recovery.

[dtango]

Hello Ravi,

First, Thank you for the response and your time. However, it would be great if you could clarify a few more things.

I'm using internal "HPE Smart Array P816i-a SR Gen10 Controller", that supports 16 HDDs.. So, I've purchased 16 HDDs (HPE 600GB SAS 15K SAS) for my DL380 ESXi server and created two RAID5 array groups, using 7+1 HDDs each (so each volume/ datastore will have 3,27Tb). 

Two questions I still have:

1. As both volumes using the same type of HDD and have the same size, I assume VMware should detect and configure the same "queue depth" for both volumes. In my case, with custome HPE image,  I'm getting 1014 for datostore1 and 240 for datastore2. Why? Is this incorrect (see below)?

I have tested original Vmware image with the same hardware and for both datastores I'v got the same parameter of "queue depth" - 1013. I still can't explain why this parameter is 1013 and not 512 or 1024 :), but at least it the same for both datastore that sound to me more logical. (see attachments in my first post)

>( 8 x 1024) / 64 = 128 single connected servers or 64 (2 x HBA) or 32 (4 x HBA) 

Can you help me with the calculation for the controller P816i SR?

2. > Changing the queue depth for QLogic, Emulex, and Brocade HBAs =>(https://kb.vmware.com/s/article/1267)

Yes, I came across to this article one day, but there is no info on how to change this parameter for P816i. Is there any way to change queue depth for Smart Arrays controllers like P840/P816i/P408, etc. ?

Will really appreciate if you could find a time and help with answers.

KR,
Aleks.