Server Crashed -- No Longer Working

[naralez]

Hi,

 

I upgraded to 6.8.2 and can't tell if I'm having a hardware problem or some type of issue related to the software. Upon restarting, the dockers won't load--so I rebooted, then the server froze. After another reboot, I was load dockers. Then, my VM wouldn't start--and then disappeared. After that, one of my HDs stopped being recognized--and all of my folders on the drive appear to be missing.

 

I uploaded the diagnostics. Any help is appreciated. I'd rate my tech skills as average--so, I may need explicit instructions if someone has ideas for me to test.

 

Thank you.

tower-diagnostics-20200207-1153.zip

[trurl]

Disk3 is disabled, but the emulated disk is mounted. Your cache is not mounted though and that is probably the reason for the other problems. You should be able to see these things on the Main page but you did not mention it. Post a screenshot.

 

Do you have another copy of anything important and irreplaceable?

 

Looks like you have CA Backup installed. Do you have backups of the important cache stuff from that?

 

SMART for disk3 has some issues. Do you have another disk to rebuild disk3?

 

You may have to reformat cache. Let's see if @johnnie.black has any ideas about that.

[trurl]

Looking at other disks, your parity1 also needs to be replaced. You should be getting SMART warnings for parity and disk3 on the Dashboard. Do you have Notifications setup to alert you immediately by email or other agent when a problem is detected? Probably had some of these issues before you actually noticed anything wrong.

[naralez]

Hi,

 

Nothing is irreplaceably important on the drives--okay if it's lost and I need to start from scratch. Yes, I can get another drive for DISK3--not a problem.

 

Screenshot of my main page is attached.

[naralez]

14 minutes ago, trurl said:

Looking at other disks, your parity1 also needs to be replaced. You should be getting SMART warnings for parity and disk3 on the Dashboard. Do you have Notifications setup to alert you immediately by email or other agent when a problem is detected? Probably had some of these issues before you actually noticed anything wrong.

I did see the drive health warnings--I was just planning on using them until they die before replacing them for a rebuild. Is that a bad strategy?

[trurl]

Might be better to rebuild disk3 without parity1 involved. Then you can either replace or remove parity1. Let's see if there are any other opinions.

 

None of these are new disks. Did you test any of these disks before trusting them in your server?

[trurl]

4 minutes ago, naralez said:

I did see the drive health warnings--I was just planning on using them until they die before replacing them for a rebuild. Is that a bad strategy?

A very bad strategy.

 

You can only rebuild as many disks as you have parity. And parity by itself cannot recover anything.

 

In order to reliably rebuild a disk, every bit of parity PLUS every bit of ALL remaining disks must be reliably read.

[naralez]

I'll replace both disks (one at a time) and then circle back--hopefully that solves the issue. Thank you.

[trurl]

Technically since you have dual parity you could replace and rebuild both at the same time. If one at a time I think I would do parity1 first since it has the worst SMART report and emulated disk3 is still mounted.

 

Be careful when replacing disks. Double check all connections, even those you aren't replacing. Disturbed connections are a frequent cause of rebuild problems.

[trurl]

48 minutes ago, naralez said:

I'll replace both disks (one at a time) and then circle back--hopefully that solves the issue. Thank you.

The rebuilds won't do anything for your unmountable cache pool.

[naralez]

5 hours ago, trurl said:

The rebuilds won't do anything for your unmountable cache pool.

How do I fix that problem?

[JorgeB]

Pool was misconfigured, one of the devices was being removed, see here for some recovery options.