On a false positive disk error the VM's shutdown?

QNAP NAS solution for server virtualization and clustering/HA/FT
Post Reply
dagoodyear
New here
Posts: 9
Joined: Thu Jan 30, 2020 6:06 pm

On a false positive disk error the VM's shutdown?

Post by dagoodyear »

Hi all,

Im currently running a TS-873 using 4.5.4.1741. I used to host all my VM's on a ESXi host with ECC memory and the bells and whistles etc.... never had a fault. If one of the raid 6 disks had a blip it continued and alerted me from the VM that ran the bulk storage systems.

This box is used to run everything from Zoneminder / OpenHab / MythTv / (many other bits) so nearly everything in the house. Usually it runs perfectly well (it's running RAID-6 / with some VM's on a separate RAID-1 SSD based volume). However....

When it thinks it's had an issue with the HD's it seems all the VM's pause on the unit (later tests seem show either the disk re-allocated or it must have been a controller blip - disk reports as in good health). This is at least the pattern I seemed to have noticed over the last few years with the unit. Assuming it has this behaviour is there a way to disable it. (If it has two disks have a fault in RAID-6 I understand it). I cant have the DB's / server VM's shutting down as it thinks fit. Is there a setting that can be set to disable this. (It's RAID-6 not RAID-0). If this is not a feature of the QNAP device's can anyone suggest why they would pause all running VM's at the random will of its choice? (It seems to happen every 3-4 months just when its not acceptable to my wife :( ).

Any suggestions as to the cause - if this is not a known feature?

Thanks in advance

David

(I really don't want to pull an old server back into use 24 / 7 / 365 days a year power usage wise, just to ensure the VM's don't suspend when not expected - worse case I'll have to use a PI as a watchdog - but that seems excessive?)
dagoodyear
New here
Posts: 9
Joined: Thu Jan 30, 2020 6:06 pm

Re: On a false positive disk error the VM's shutdown?

Post by dagoodyear »

The disk that flagged up the odd fault was changed, with a fresh unit - even although it passed all tests. Since this the pausing of VM's has not occurred, in case anyone has the same issue.
Post Reply

Return to “Server Virtualization & Clustering”