VASA Provider HA Soloution

[paymanzox]

Hello everybody

We Used vvol in Storage Configuration

We face a challenge at work. If there is a VASA Provider Service  failure, there is a solution for example HA?.

Thank you 

[veeyarvi]

Hello Paymanzox,

The VASA failure normally does not affect the already running one's IOs. But you cannot do any modifications to these. 

[paymanzox]

hello dude

You mean, if you come up with a problem for vasa, you can still take the example Snapshot. Of course this is unlikely

[veeyarvi]

Hello Paymanzox,

No, I did not mean that. Apologies if I was not clear.

What I meant is, even though the VASA is down, the IO will continue for any already mounted volumes. But we cannot perform any other operations related to those volumes (of course we cannot take a snap as well). Even we cannot unmount them. 

[paymanzox]

hello dude

Thank you very much for your description.

But my question still remained unanswered. Is there a solution for VASA HA?
I know some brands like "IBM" have a solution to this issue

thank you

[veeyarvi]

Hello Paymanzox,

HPE 3PAR VASA provider also has High Availability. It allows the VASA provider to run on an HA cluster, stretching across all nodes on an HPE 3PAR array. This provides for continuous VASA provider uptime even if multiple components fail on an HPE 3PAR array.

A high-level failover/failback scenario is described in page 24 of HPE 3PAR StoreServ and VMware VVols
Replication Implementation Guide (https://h20195.www2.hpe.com/V2/GetPDF.aspx/a00008302ENW.pdf)

Hope this answers your query.

[paymanzox]

Hello veeyarvi;

I saw the document you mentioned. I can say that this solution exists within Substrate vmware and relates to solutions such as SRM. It is either provided by the company HPE through the "peer persistence" for Metrocluster solution.

Anyway, thank you very much for the time you answered

‌best regards