Media directory nuked itself? (Solved...?)

[sumidor063]

Hey all,

 

So... not sure how I did it, but I managed to nuke my media directory, and now everything is broken. In the Unraid GUI (I'm now realizing that this was probably my mistake, as I've done the following via Krusader numerous times before without any issue), I went into my cache pool, then appdata/plex/Library/Application Support/Plex Media ServerCache/PhotoTranscoder, with the intent of deleting everything inside of it to clean things up. Clicked the select all button, delete, yes, and it appears to have deleted far more than just the photo cache directories.

 

Now, my media folder (which is where all of my media is kept, namely for Plex), is showing up as nothing. I don't know what I did wrong, nor how to fix this, and my attempts to search for a similar issue/resolution have resulted in dead ends. Any help would be appreciated.

 

Running Unraid 6.10.3.

 

 

Edit1: None of the files within media appear to have been deleted, as there is no disk activity and my usage levels haven't changed. I also just went and turned off Docker, and noticed the following. I'm now afraid to try to turn it back on. Both the .img within that path and the appdata directory still exist and appear completely intact.

 

Edit2: This might be helpful. But I don't know what it means.

 

 

t1-diagnostics-20221001-0319.zip

Edited October 1, 2022 by sumidor063

[sumidor063]

Since my most recent edit (14 minutes ago at the time of this reply), I decided... y'know what, if everything is gone and I have to start from scratch, rebooting can't hurt things worse than they already are. I have since done so, and everything appears to be back.

 

If anyone wants to dig through to put the pieces of the puzzle together to satisfy your curiosity (and mine, admittedly), I'd happily listen to ideas. The diag attached to the original post is pre-restart, and the actions I took step-by-step were noted in the original post. Today's lesson: don't do things differently just to fiddle, when you know another method works just fine.

[JorgeB]

1 hour ago, sumidor063 said:

rebooting can't hurt things worse than they already are.

It's the way to fix this when it happens:

Oct  1 03:12:29 T1 shfs: shfs: ../lib/fuse.c:1451: unlink_node: Assertion `node->nlookup > 1' failed.

It's this issue: