Rootfs is getting full


elco1965
Go to solution Solved by trurl,

Recommended Posts

Fix common problems was reporting "Rootfs getting full" I mostly followed the directions that lead me to the creation of this post. I have been having a couple of issues lately. One of them being with the web GUI. It would become unresponsive. I could log on but when I clicked on the Main tab the drive assignments including the cache pool were missing. VMs would also disappear. The only way to get these back was to stop and restart the array via the terminal. 

 

I first noticed the issues when attempting to preclear a couple of drives. The preclear would be aborted almost immediately fail at the beginning of the process with the error "detected memory low". I read through e few post, some suggested bad memory. 

 

I also found the "Where can I see what folders are taking up my RAM?" I ran the script "https://raw.githubusercontent.com/Squidly271/misc-stuff/master/memorystorage.plg" and it returned /mnt/plex....something. I unfortunately didn't save the output before the server became mostly unresponsive. The strange thing about the "what folders are taking up my RAM" script, Plex and all other docker containers were stopped and the dashboard was reporting about 70% of the RAM was in use. 

 

I had some extra memory and swapped out some of the modules. I am 19 hours into a Memtest with no errors reported. The server has ECC server RAM. I read through the forums and some have suggested the version of Memtest that comes with unRAID is obsolete or doesn't work with ECC memory. I don't know if I should continue with the test or not.

 

Any help with this would be greatly appreciated. Thank you

blackhole-diagnostics-20221120-2247.zip

Link to comment
8 hours ago, elco1965 said:

Is there a good reason to continue with the Memtest on ECC RAM? 

If you have ECC RAM then you should download the latest memtest from memtest86.com as it can test ECC ram.  The version that comes with Unraid is an earlier version (because of licensing issues) that cannot test such RAM.

  • Like 1
Link to comment
12 hours ago, trurl said:

Filling rootfs will definitely make for an unresponsive server, since the OS has no room to work in if you fill rootfs

/mnt/plex is a path in rootfs. Check your docker host paths.

 

Also, your system share has files on the array

 

Thank you sir. This was it.

I moved the system files from the array and fixed the plex container mapping. Now when I run the "Where can I see what folders are taking up my RAM?" found here 

it returns the attached and looking much better. I still 2 files that show up as /mnt, 0 /mnt/rootshare and /mnt. I beleive that /mnt/rootshare was created following Spaceinvader One's video about root shares for Windows users. I am guessing it's supposed to be there and looks as if it is using zero memory. I don't know about /mnt Also, when browsing the files using the file browser, there are 2 folders in /root I do not know what they are. "disks" and "remotes". I tried to delete disks but it returns on it's own. I am assuming these are supposed to be there or were created by mistake?

Folders .png

Memory Output.txt blackhole-diagnostics-20221122-0849.zip

Link to comment

/mnt is in rootfs, but it is supposed to be there as a place for things to get "mounted". All array disks, pools, and user shares are mounted in /mnt

 

/mnt/disks, /mnt/remotes, /mnt/rootshare are created by Unassigned Devices as places for it to mount unassigned disks, remote shares, and the rootshare that includes all user shares.

 

Anything else that specifies a path in /mnt, such as /mnt/plex, will create that path, but with nothing getting mounted there, anything stored there is taking up rootfs space.

Link to comment
21 hours ago, trurl said:

What is the purpose of your cache:prefer share n-------d ?

 

21 hours ago, elco1965 said:

I am not seeing it under the shares tab.

The diagnostics are anonymized so I don't know the exact name of the share. Diagnostics shows you have a share beginning with n, ending with d, 9 characters, set as cache:prefer, and with files on cache.

 

Post a screenshot of User Shares page

 

Link to comment

This reminds of when my Grandfather would slap me upside the head for saying or doing something stupid. 

 

You are correct. I was stuck on a folder that I saw in /mnt. This was one of the things I deleted the other day. 

 

Anyway, I was having a couple of issue with Nextcloud as well. I put that share on the cache hoping for a gain in "performance". I will be moving it back to "yes".

 

Thanks again for you're help. As someone who is really out of his league when it comes to the proper use of unRAID, I really appreciate the community here.  

Link to comment

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.