Had to rerun SATA cables to cache drives, lost VMs in domains only??

[btrcp2000]

I'm not heartbroken because my VMs were a mess anyway, but now that the important ones are up and running better then ever I don't want this happening again.  Why would one folder have gotten blown up?  It's as if there was never anything in the domains folder, and there were 6 VMs the other day.

 

What is the best way to backup VMs? I'm seeing lots of options, currently I just have Crashplan syncing that folder, but I'm guessing that's not a good idea because it doesn't know to shut them down?

 

 

[Squid]

Quote

Had to rerun SATA cables to cache drives, lost VMs in domains only??

You've got another underlying problem where a drive is missing or the like...  Post ze diagnostics

[btrcp2000]

unraid-diagnostics-20191129-1500.zipSorry, I should know better by now. Attached.

 

That rings a bell though. On first boot after re-running cables, it lost one of the cache drives.  On another boot, the missing drive showed as unassigned, so I assigned it as cache thinking it would just sync up with the other cache drive, as that's the point of having two (right...?)

 

Working fine now showing dual cache drives. Still need to learn best practices to offsite backup (this weekend's project)

[JorgeB]

11 minutes ago, btrcp2000 said:

so I assigned it as cache thinking it would just sync up with the other cache drive,

In this case Unraid would consider it a new device, there would be a warning "all data on this drive will be deleted at array start", there are ways of re-adding without deleting the data, but not this way.

[btrcp2000]

But what would have happened to the data on the other drive that never showed any issues? Dockers survived, so why just the domians folders?

[JorgeB]

Doing that can corrupt the pool, also depending on when the pool was created it could have been non redundant, because of a bug with v6.7, but would need diags to confirm.