Data Loss After Hard Shutdown

[jekupka]

I'm going through some growing pains learning the ins and outs of the OS and it's applications and unfortunately have had to do a hard shutdown a couple of times recently. Each time the array is brought back up after the shutdown I have lost all data acquired since the last parity check. Since this has happened each time I am now assuming that I risk losing all data obtained between parity checks if a hard shutdown is performed. Is this thinking correct? If not, why would I lose data if the disk is still in the array?

unraidinblood-diagnostics-20230119-1532.zip

[itimpi]

You should not be losing any data so something else must be going on.

[trurl]

Parity doesn't contain any of your data.

 

Jan 19 14:37:25 unRaidInBlood  emhttpd: unclean shutdown detected

Unclean shutdown will start a non-correcting parity check when you restart, and often results in a (relatively) few sync errors. I suppose it could also lose some writes to data disk but only writes that had not been flushed yet. None of this is anything like what you describe and parity checks don't affect your data disks in any way.

 

Can you tell us more about what exactly you are doing? Sometimes the words "parity check" are used by new users imprecisely and could possibly mean many things to them. What does it mean to you?

 

Do you ever get disabled disks? Or unmountable disks? Did you ever do New Config? Have you formatted any disks that were already in the array and had data?

 

Jan 19 14:40:12 unRaidInBlood kernel: mdcmd (36): check correct
Jan 19 14:40:12 unRaidInBlood kernel: md: recovery thread: check P ...
Jan 19 14:40:12 unRaidInBlood kernel: md: recovery thread: P corrected, sector=0
Jan 19 14:40:12 unRaidInBlood kernel: md: recovery thread: P corrected, sector=8

Looks like it started a correcting parity check after reboot and it was correcting so many sync errors it quit logging them. Did you tell it to do a correcting check?

 

Are you sure you ever did the initial parity build?

 

Looks like you might be trying to use USB connections for some of your drives. USB connections are NOT recommended for array or pool disks for many reasons.

 

[jekupka]

So after leaving the situation for a bit to clear my head I realized two things were happening at the same time one of which I didn't think would be an issue. Issue one was the unexpected restarts due to a self created issue. What was taking place at the same time was the deletion of an old Media folder which I did NOT know was being used by Radarr for "Collections". It seems that by deleting the root folder Collections was using it also deleted the files associated with that collection even though they were located in a different root folder. I'm waiting for the parity check to finish before I reacquire the missing data and verify my hypothesis, but nothing else logically makes sense. 

 

Thanks to those that took the time to look and offer input.