I came across an error today on one of my disks where it stopped working all of a sudden.  It looks like it is now disabled.  Not sure how to tell what is wrong with it.  I will post my diagnostics.  finalizer-diagnostics-20190413-2119.zip

36 minutes ago, mgranger said:

Not sure how to tell what is wrong with it.

If you look on the Dashboard, you should be getting a SMART warning for that disk. If you mouse over it should tell you something about it. You can also click on the disk in Main to get to its page and look at the attributes.

 

The disk has one pending sector. It's not clear if that caused the disk to be disabled or something else. 

 

Your syslog seems to indicate you rebooted with the disk already disabled, and it is currently doing a correcting parity check. Is that correct? We can't see what happened before the reboot to cause the disk to be disabled.

 

Not related but you have a 40G docker image that is filling up. If you have your docker applications configured correctly it is very unlikely you would ever need even 20G, and making it larger than 20G doesn't fix anything, it just makes it take longer to fill.

 

 

Here is a screenshot of the attributes. yes i know the docker size is probably too big but i do have issues with it filling up. not sure what i have set wrong. i use home assistant and influxdb and mariadb which are the only ones that i can think of that may grow so i am sure it is one of those. 

 

Edited April 14, 20197 yr by mgranger

9 hours ago, trurl said:

Your syslog seems to indicate you rebooted with the disk already disabled, and it is currently doing a correcting parity check. Is that correct?

Did it complete the parity check? It looked like it was correcting a lot of errors on parity2. If it has completed, what was the result?

 

Why was it doing a parity check anyway?

 

Post a new diagnostic.

Yes it completed the parity check.  I started it.  

Not many errors then, about what might be expected from an unclean shutdown. I didn't notice an unclean shutdown in syslog so maybe it happened earlier.

 

You can rebuild the disabled disk to itself, or you can rebuild to a spare disk. The advantage of using a spare is you have the original disk as a backup if there is a problem rebuilding.

 

The other disks looked OK, but you should double check all connections before rebuilding.

 

Do you want to use a spare disk or rebuild to the same disk?

 

 

I am sure there was an unclean shutdown earlier.  I don't really have a spare disk at this time to rebuild to so probably the same disk.  I don't mind buying one if this is going keep happening with this disk.  I honestly don't have anything real important on this disk at this time so I am not too worried about losing anything.  I have some backups of VMs and my photo database but they are backups so if I need to redo I should be ok.

1 hour ago, mgranger said:

I honestly don't have anything real important on this disk

The problem with that statement is that unraid parity requires ALL remaining disks to be perfectly healthy to accurately reconstruct a failed disk. So, if you were to have one of your disks that DOES have some important unexpectedly fail, unraid would be unable to rebuild it if this dodgy disk doesn't perform.

48 minutes ago, jonathanm said:

The problem with that statement is that unraid parity requires ALL remaining disks to be perfectly healthy to accurately reconstruct a failed disk. So, if you were to have one of your disks that DOES have some important unexpectedly fail, unraid would be unable to rebuild it if this dodgy disk doesn't perform.

Ok so what do I need to do to rebuild the disk.  I was thinking that since I had 2 parity disks I was able to have 2 disks go bad at the same time and still be able to rebuild them without losing anything.  I have yet to have to rebuild a disk yet so I am not quite sure the correct way to do this.

To rebuild to same disk:

1. Stop array
2. Unassign that disk
3. Start array with disk unassigned
4. Stop array
5. Reassign that disk
6. Start array with disk reassigned to begin rebuild

4 hours ago, trurl said:

you should double check all connections before rebuilding.

don't forget this. Maybe not a problem but we don't really know what caused the disk to be disabled.

the rebuild seemed to complete fine. everything seems to be back to normal. will just be sure to keep any eye on it. thanks for the help

1 hour ago, mgranger said:

will just be sure to keep any eye on it.

Unraid will keep an eye on it. You should setup Notifications so Unraid can let you know.

2 hours ago, trurl said:

Unraid will keep an eye on it. You should setup Notifications so Unraid can let you know.

They are setup.  This is how I knew it was disabled the first time.  Also I know this is a little off topic but you mentioned my docker image being 40G and that 20G should be plenty.  I went through and looked at which containers were taking up the most space and it looks like home assistant is taking up 3.17G and resilio-sync is taking up 6.63G.  Everything else seemed fair size wise.  Any thoughts on what might setup wrong with these?  I would like to get them corrected if possible.

You might just try deleting and recreating it at 20G then you can add your dockers back exactly as they were using the Previous Apps feature on the Apps page.

 

The way docker image gets filled is some application writing to a path that is not mapped to Unraid storage. Typical mistakes are not using the same upper/lower case as in the mappings, or using a relative path instead of an absolute path (beginning with / ). You usually need to fix this within the application itself, not the container settings.