There may be some down time in our future

[InsaneJ]

Last night one of the hard drives in the server failed. It's part of a raid-1 array on which the server's OS, websites, databases, e-mail and some other things are stored.

The drives were ordered June 1st 2014 so they're a little over a year old. They're 3TB Seagate drives. And in hindsight I should never have bought Seagate drives because they are notoriously unreliable. But yeah. Hindsight 20/20 right? :)

Anyway, the drive will need to be replaced. For this the server doesn't need to go down. But because I don't have a fancy (and very loud) server rack but a silent PC case I will need to move it around a bit to swap drives. Moving around spinning disks is never a good idea so for that I will probably shut down the server.

Running everything on a single Seagate drive while we wait for our RMA... Or I could order two new non-Seagate drives. Choices, choices :B

[Sverf]

Should reboot it anyway tnite, last night the reboot failed and now the host OS is unreachable :-/

Best to run shutdown in each of the vm's and then just reset the server I think, unless you want to hook up a monitor and keyboard to see whats wrong in the host OS :p

So maybe after that reboot it will pick up the drive again, prolly not.

[InsaneJ]

The drive didn't come back to life so I've removed it from the server after we had restarted it about 15 minutes ago. I've ordered a new drive since RMA can easily take more than a month and I don't want to risk running on a single Seagate drive.

The new drive will go into the server to make the raid array whole again. I was about half an hour too late ordering a new drive so it should arrive on Wednesday instead of tomorrow. The drive that eventually comes back from RMA will go into Sverf's server and act as storage for backups from both servers.

[Jiro_89]

Thanks for the update :)

[InsaneJ]

So I replaced the drive tonight.
Then the power supply died... again >_< I had just replaced it a few weeks ago.
Mupje graciously offered the power supply from her PC so now at least the server is up and running again. And it's running on a Seasonic PSU so I'm not expecting any more problems there.
And lastly I just ordered a new power supply for Mupje.

I had hoped to spend my night on the couch. Oh well. Maybe next time ;)

Bonus pic for those interested. This is what the raid storage manager of the server looks like:
2227

[Sverf]

Nice one, just got the email that the rebuild of the OS mirror is complete!

So in exchange for the power supply you have to sleep in bed? :p
You probably meant evening on the couch :cool:

On another note, I've switched java on the server from 1.7 to 1.8 yesterday as well, hope everything keeps on doing what its supposed to after the daily restart today.

[Marius49]

That storage manager looks so exotic :cool:. Let's hope it's not a hint for exotic temperatures of the drive ;o.

[InsaneJ]

The new drive is spinning at a comfortable 33 ºC (91.4 ºF) Tropical temperatures are the way to go :)
2231