Array will no longer start

[dradiation]

Hi,

I am having issues with my array no longer starting. When I click start it just hangs indefinitely. I am able to start in maintenance mode, and I ran xfs_repair on both of my disks, but am still unable to start the array. I am almost certain that my docker image was corrupted as I was working with docker when I last was able to start the array and the docker service keeps giving a failed to start error. I have the docker service currently disabled. I have tried rebooting unraid server times to no avail. Any suggestions would be greatly appreciated! attached are the diagnostics, following a reboot after an unsuccessful array start.

archipelago-diagnostics-20230412-1703.zip

[JorgeB]

Enable the mirror to flash option in the syslog server, start array and post that.

[dradiation]

Here are the diagnostics with mirror to flash enabled.

 

 

archipelago-diagnostics-20230413-1735.zip

Edited April 13, 2023 by dradiation
Updated logs, after it spent an hour trying to start array

[JorgeB]

The persistent syslog is save in the flash drive /logs folder, you need to upload that separately.

[dradiation]

Shoot apologies, ok here is the syslog file from the flash drive.

syslog

[JorgeB]

Unfortunately there's nothing logged that I can see:

 

Apr 13 15:23:13 archipelago  emhttpd: Mounting disks...
Apr 13 16:54:40 archipelago  shutdown[15113]: shutting down for system halt

 

Since you don't have parity you could try starting with just disk1 or just disk2 to see if there's any difference.

 

[dradiation]

OK so, I ran a read-check all day it it found drive errors. Is there any way to recover from this? Would the xfs_repair be able to help?

syslog(1)

[JorgeB]

Run an extended SMART test on disk2, xfs_repair cannot help if the disk is really failing.

[dradiation]

OK, I ran SMART on it and it kicked back with a read failure. So I take it the drive is failing and to scrap it?

archipelago-smart-20230416-0937(1).zip

[dradiation]

OK I was (weirdly) able to start the array right after posting the previous comment so I am going to pull the data off. From the SMART data giving a read fail I assume my best course of action is to remove/replace the drive and make sure I get a parity drive going?

 

Edit to add: Is there a way to have unraid move the data from disk2 to the functioning disk1 to keep the docker appdata and everything? I know how to do it through the terminal, I guess I am asking is: will unraid still be able to find everything if it is moved from disk2 to disk1 or is there an inbuilt functionality to do the move and keep any mappings correct?

Edited April 16, 2023 by dradiation
extra question

[JorgeB]

15 hours ago, dradiation said:

From the SMART data giving a read fail I assume my best course of action is to remove/replace the drive and make sure I get a parity drive going?

Yes.

 

15 hours ago, dradiation said:

Is there a way to have unraid move the data from disk2 to the functioning disk1 to keep the docker appdata and everything?

Only manually, and if all everything is using /mnt/user it will still keep working.

[dradiation]

Ok I will get everything moved manually then and get a new drive in. Thank you for the help!