heelphan Posted March 5, 2020 Share Posted March 5, 2020 (edited) Hello, Hopefully somone can help me, I'm kind of panicking a bit. Been running Unraid for about 3 months now. So I was working on my Windows VM about 40 mins ago and everything was fine and then I went to open my Sonarr Docker. It froze and didn't want to open. So, I restarted it. Then when I opened it I didn't see any of my added series, but instead saw the main splash screen with SQL errors over and over. I tried opening my Plex docker and it wasn't accessing my libary, all of my libraries had exclamation points by them, like it couldn't locate the libraries. So, I shut down my array and started it back up again. Of course I was kicked out of my VM as it restarted. The Dockers spun back up, but now every single Docker I have is acting like it is brand new and has never been configured - Sonarr, Plex, Deluge, etc. I then tried my Windows VM and it wouldn't start giving this error: "Cannot access storage file '/mnt/user/domains/Windows 10/vdisk3.img': No such file or directory" My cache drive is a 1TB Crucial M500 that is only about 2 months old. This is where the appdata for all dockers and Windows VMs are stored, which makes me think something must be wrong with my Cache drive. Luckily I do run CA Backup to backup my appdata folders to my array. On a hunch I did take a look at my SMART data for my Cache drive and there are 2 items in "Pre-fail" status: The RAW Read Error rate and the "Unused Reserve NAND block". Could my cache drive be failing? I am attaching my full diagnostics zip file. Could there be any security breaches happening? I see a couple entries about "device vethf41a825 entered promiscuous mode" around the time I lost access to everything. Does anyone have any idea what is happening? tower-diagnostics-20200305-1426.zip Edited March 5, 2020 by heelphan edit subject title Quote Link to comment
heelphan Posted March 5, 2020 Author Share Posted March 5, 2020 By the way, I have shut down the server until I can get home from work on the chance that it is a security breach. Quote Link to comment
heelphan Posted March 5, 2020 Author Share Posted March 5, 2020 I little more info. Not too long before this happened I installed a docker called "Unmanic" and pointed it to a drive with many GBs of files. I also created a folder called "Unmanic" directly on the cache drive as per the config said to do. The container did start up but perhaps I did something to my cache drive when that happened? Possibly my cache drive filled up or my docker.img file became too large? My problem seems similar to THIS thread. Thanks for any help anyone can provide! Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 Ok, powering the server back on when I came home, my cache drive now has lost the 2 VM images it had on it, but my appdata folder still says it is on there. However, when I start up a docker container, like Plex, it’s like it is a brand new container and doesn’t see my appdata. I ran a quick SMART test on the cache drive and no errors were found, but the two previously mentioned things are still in “Pre fail” status. Quote Link to comment
JonathanM Posted March 6, 2020 Share Posted March 6, 2020 Without starting the container, can you go to the edit screen for unmanic and post a screenshot showing your paths? Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 Unfortunately, I deleted Unmanic after I saw it was not going to be something I wanted to use. I mapped the "Movies" drive to a folder where I keep I movies I'm planning to convert, its a folder on my array only, not on the cache, that has probably 500 GB worth of data. Then for the /tmp directory, I mapped it directly to my /mnt/cache/tmp and created a folder called Unmanic, so the full path was /mnt/cache/tmp/Unmanic. In my panic of re-tracing my steps I navigated to that folder in midnight commander and deleted it. Quote Link to comment
Squid Posted March 6, 2020 Share Posted March 6, 2020 There will be a file on the flash drive: /config/plugins/dockerMan/templates-user/unmanic.xml Post that. Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 OK, got it, see attached. my-unmanic.xml Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 (edited) Do you think I should just Follow the steps in the CA restore/backup plugin to restore my last Backup of the appdata folder? Or is there anything else that can be done? Also, I’m wondering what happened to my VM images that were stored on my cache drive in the Domains folder? My Win 10 VM image was 250GB. Now my cache drive shows only 40 GB used out of 1TB. It appears it’s all of my docker data, but again, my dockers appear to be brand new and aren’t able to access that data apparently. Edited March 6, 2020 by heelphan Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 Woke up this morning to see that the Cache drive now reads only 7.85 GB used instead of the 40GB it said yesterday. I do see according to the log that TRIM ran at 2AM: Mar 6 02:00:01 Tower root: /etc/libvirt: 920.6 MiB (965349376 bytes) trimmed on /dev/loop3 Mar 6 02:00:01 Tower root: /var/lib/docker: 9 GiB (9659695104 bytes) trimmed on /dev/loop2 Would this be a further indication that my Cache drive is going bad? I'm at a loss here, sorry for so many posts! Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 Latest diagnostics attached. Thank you for any help! tower-diagnostics-20200306-0948.zip Quote Link to comment
trurl Posted March 6, 2020 Share Posted March 6, 2020 5 hours ago, heelphan said: Mar 6 02:00:01 Tower root: /etc/libvirt: 920.6 MiB (965349376 bytes) trimmed on /dev/loop3 Mar 6 02:00:01 Tower root: /var/lib/docker: 9 GiB (9659695104 bytes) trimmed on /dev/loop2 These are the virtual disks for docker and libvirt images. Some of your system share is on disk1 though, so that could be where your 20GB docker image is. Some of your shares are cache-yes, so maybe those got moved to the array. You can see how much of each disk each user share is using by going to Shares - User Shares and clicking Compute All. Didn't see anything obviously wrong in that docker template. Can't vouch for the docker itself though. I see the author says it is beta, and CA shows that in its listing. Are you sure you didn't reformat cache? That seems the most likely explanation. Quote Link to comment
heelphan Posted March 6, 2020 Author Share Posted March 6, 2020 I don't think I reformatted the cache, as all the appdata folders were still there, just without any of their data. I don't know what happened but I bought a new SSD just to be safe and did a restore of my CA Backup from a few days ago and deleted my docker.img file and recreated it then reinstalled all my dockers from CA and now everything is back to the way it was before. The only thing I have completely lost, it appears, is my Win 10 disk IMG from the domains folder. That sucks, but it is able to be recreated at least. In the end I have no idea what could have happened. I don't know if my old Cache SSD is bad or not. Quote Link to comment
trurl Posted March 7, 2020 Share Posted March 7, 2020 16 hours ago, heelphan said: I don't think I reformatted the cache, as all the appdata folders were still there, just without any of their data. If you reinstalled/restarted your dockers on a formatted cache then they would have created new appdata folders. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.