OVC 5 crash

[Semmaris]

Hello,

On our 4 hosts cluster federation, 2 OVCs version 5 crashed today. 

One the morning the second in the afternoon with the same fatal error 

We don't know the cause of the crash but we must force reboot the OVC and the SVT service did not restart because of a norestart file. (which seems to be logical)

1st time for us so we had to had the HPE Support which was great.

My advice, in case you have an OVC which crashes, call ASAP HPE Support to manage the case.

[s-]

Hi @Semmaris 

As per screen shot error "AIO OR io_getevents" message, it looks to me SVTFS is   crashing due storage sub system issue which could be due to Controller, Controller Firmware/Driver, Drive, Drive firmware/Driver etc However it is best to engage the support ASAP to find out the exact root cause of the issue.

Hope it helps.

Regards,
Jaipal

[Semmaris]

@s- Same crash this morning

Remove of the norestart file, force restart the OVC and it's ok.

But now my infrastructure is unstable, i think a HPE/Simplivity engineer should look at he logs to help us find some stability.

[Semmaris]

Another crash of OVC this morning with dataloss.

This new 5 version is really problematic for me.

[gustenar]

@Semmaris 

Don't just restart the svtfs service, have support investigate the reason of the failure. Restarting the service without proper investigation is not going to fix it for you and it's just going to keep crashing. 

[Semmaris]

I know but this morning i had vms stuck and dataloss, the priority was to restart my node.

[gustenar]

When this occurs contact support immediately as more damage can be done without proper troubleshooting. Depends on every scenario but your data might still be available and retrievable.