HPE OneView Temporarily Unavailable

[Eric_RP]

Anyone else have their OneView instance go into a Temporarily unavailable state and not allow anything to be accessed? Ours is currently doing this and we can't access anything beyond the initial dashboard. The OneView VM has 4 vCPUs and 12GB of vRAM. VMware performance stats show 25% CPU and 38% Memory usage. 

When this has happened in the past we've tried rebooting the OneView VM but the issue comes back a few days later. We have 10 enclosures, 120 servers, and 117 profiles in this OneView installation.

I tried to open a support case for this a while back but it went nowhere. Hopefully one of y'all has seen this before and can offer some suggestions. Thank you!

[ChrisLynch]

Can you please provide what version of HPE OneView you have deployed and are using?

. The OneView VM has 4 vCPUs and 12GB of vRAM. VMware performance stats show 25% CPU and 38% Memory usage. 

We do not support adding more resources to an appliance than what is defined in the OVA.  Adding more resources will not help scale the appliance.  In your case, I suspect that the issue is disk IO related.

I tried to open a support case for this a while back but it went nowhere

Support cases should always be driven to closure, and not just by a support engineer closing the case.  Can you PM me the support case ID?

[Eric_RP]

Curiously the OneView appliance started behaving correctly a few minutes after I hit the Post button. Its like it knew.  

 OneView 2.0. Firmware field says 2.00.02-0219211.

The Installation Guide, Help Guide, and Support Matrix mention this regarding the VM requirements: 

The appliance VM requires the following:
• Two 2 GHz or greater virtual CPUs.
• 10 GB of memory dedicated to the appliance.
• 170 GB of thick-provisioned disk space.

Based on the first line I thougth increasing the vCPU from 2 to 4 would help things. There is no mention of adding resources not being supported. If that is the case you guys might want to look at updating the documentation to reflect that. 

Disk I/O looks great from the VMware side. Disk latency has averaged 1ms or less and the highest spikes have been 7ms when the appliance is backed. up at night via the New-HPOVBackup OneView PowerShell command. This VM was originally housed in a big 3PAR storage array. When we first started having these issues we migrated it to an isolated host with local 15k drives. When we continued having issues we migrated it to a different host using DAS. In all three storage platforms and different VMware hosts the issue has reared its head eventually. 

SR#: 4653163658 back in January. I can't pull up the case details since its bene more than 30 days.

Like I said, its working now, but It would be good to know what to do in the future when it decides to happen again beyond just rebooting and hoping for the best.. At this point I'm thinking it may be weorthwhile to deploy a new appliance and restore from backup. I'm not sure what impact that might have.

Thank you for any insight you may be able to provide!

[splatt · Accepted solution]

I had the same issue - Onview running on a dedicated ESX host so plenty of resources

Apparently it is a known issue with Oneview being upgraded from a previous version to 2.0 which is what I did

This leaves numerous old tasks runnning unsuccesfully apparently, and you must log a call and get HP support to remote onto the appliance and run a script on the appliance.   This worked for me.