Nothing new

[BoltSnap]

Yes. The server crashed.

it was restored from a backup copy

Don't you have the server hosted at a big name cloud service?

 

[Wookie]

Don't you have the server hosted at a big name cloud service?

Yup.

 

[jborg]

Could we possibly have some more details? As alluded to, since it's cloud-hosted, it's unlikely the underlying hardware/VM disappeared and necessitated a restore. Among the remaining options are mistakes in administration causing data loss (been there, done that, it happens) and external breaches (hacker exfiltrating all our account credentials, requiring wipe and restore from before the breach plus fix of the initial problem). I'd like to know that it's not the latter...

 

[jborg]

it's unlikely the underlying hardware/VM disappeared

... though not actually as unlikely as I'd have assumed.

[Most] Amazon EBS volumes are designed to provided 99.8%-99.9% durability with an AFR of between 0.1% - 0.2%,

(Amazon EBS Features - Amazon Web Services)

That's a one-in-500 to one-in-1000 risk of losing a volume in any given year, which honestly seems like quite a lot. (Though still unlikely to be the cause of any given specific outage.)

 

[CamC]

Could we possibly have some more details?

Site Support: Server restore Dec 29

Due to a failed upgrade, a backup from the 28th was used to restore the server. As a consequence, all posts from the last 15 hours have gone.

Shame there wasn't a backup taken right before the upgrade. Or maybe there was, and it wasn't restorable, it happens.

 

[BoundForElsewhere]

This whole situation kind of reminds me of my own mental capacity. I know I did something yesterday, I just can't remember or prove it.

Oh well. I'm not in jail or divorced so all good.

 

[jborg]

Ah, good, nothing to worry about then.