August 8, 201510 yr I've tried to install a new docker, and also to update sickrage within an existing docker and get this error for both. My docker.img file is 30GB and when I ran docker images, I got: REPOSITORY TAG IMAGE ID CREATED VIRTUAL SIZE binhex/arch-sickrage latest 816cb9f461d1 5 days ago 713 MB gfjardim/crashplan latest 4abc20213697 3 weeks ago 637.3 MB gfjardim/owncloud latest e432c58fda24 7 weeks ago 510.6 MB needo/deluge latest 60e9118749e9 7 weeks ago 367 MB needo/plex latest 8906416ebf13 3 months ago 603.9 MB <none> <none> c9813f09f163 4 months ago 215 MB This doesn't seem to be 30GB. I check /mnt/cache because that is where my docker.img resides and it has 120GB left. I had this happen before, so I removed the docker.img and rebuilt it, reinstalling all my dockers. However it has happened again. How do I fix this? thanks david
August 8, 201510 yr just double checking your docker.img is set to 30G and not 3G If you add up your VIRTUAL SIZE of your docker images it is : 3046.8 MB which is 2.9754 GB If your docker image is 3GB instead of 30 GB i can see this no space left on device error happening Goto Settings -> Docker and in the Docker Volume Info box, what does the devid line read? it should look something like this devid 1 size 20.00GiB used 9.04GiB path /dev/loop0
August 8, 201510 yr Author Label: none uuid: e87b6ba7-aa22-4a4f-8ab0-61cf28f7ab97 Total devices 1 FS bytes used 28.28GiB devid 1 size 30.00GiB used 30.00GiB path /dev/loop0 btrfs-progs v4.0.1 so it thinks it is full as well? david
August 8, 201510 yr Yes. If your dockers arent taking up that much space, perhaps you are downloading content to the docker img instead of the cache drive??
August 8, 201510 yr Author Does this help? Filesystem Size Used Avail Use% Mounted on /dev/loop0 30G 30G 4.0K 100% /var/lib/docker looking in /var/lib/docker/btrfs/subvolumes/* I see many (100's) of directories with long numbers as the directory name and each is 100s of MB large. /var/lib/docker/btrfs/subvolumes/02b99f0f04903f89daf2ea9027919dea120354e2e6e576286b43701ebb4254ed bin/ config/ dev/ home/ lib64/ mnt/ proc/ run/ srv/ tmp/ var/ boot/ data/ etc/ lib/ media/ opt/ root/ sbin/ sys/ usr/ I tried to stop docker, but it hung, and I can't shutdown tower because it is just sitting there, for 10 minutes, trying to unmount the disks. david
August 8, 201510 yr are you docker mappings set to use a share or cache? if your mappings are set incorrectly you could be using the docker image instead of a share
August 8, 201510 yr You'll need to delete yourdocker.img and recreate from your templates. Once your docker.img is full you have no means of telling which files may have been corrupted via truncation. I suspect you do not have your directory mappings set correctly between host and docker and that some docker gobbled up all the space it could. Likely candidate is owncloud and your downloading programs. The other possibility is you went through extensive docker install uninstall reinstall or docker program updates that filled all ypur space.
August 8, 201510 yr Author All my mappings are to /mnt/cache/apps/* Things have been running without issue for months. Not sure what happened to break things. I had to rebuild all my containers two days ago because it was out of space, and now again. I've not added/changed containers. I'll recreate and watch the docker.img size as I do so. I only update one container (sickrage). I did try to download the Google Music Manager docker manuall, maybe that is what screwed it up. During that pull I got the out of space error. david
August 8, 201510 yr Author So after installing my dockers I ran the scrubber: WARNING: errors detected during scrubbing, corrected. scrub device /dev/loop0 (id 1) done scrub started at Sat Aug 8 10:36:51 2015 and finished after 338 seconds data_extents_scrubbed: 75459 tree_extents_scrubbed: 140933 data_bytes_scrubbed: 2867515392 tree_bytes_scrubbed: 2309046272 read_errors: 0 csum_errors: 0 verify_errors: 0 no_csum: 288 csum_discards: 8971 super_errors: 0 malloc_errors: 0 uncorrectable_errors: 0 unverified_errors: 495 corrected_errors: 0 last_physical: 6211764224 are errors expected? david
August 8, 201510 yr Errors are NOT expected. However they seem to be abundant with the BTRFS docker.img when it fills up.
August 8, 201510 yr Author Odd I get errors on the brand new img and only 5GB of the 30GB are in use. david
August 8, 201510 yr Author This isn't good. I left for an hour and the docker.img is now 11G used instead of the 5G when I left. I have done nothing to it. Anyway I can try to figure out why it is filling up? david
August 8, 201510 yr You likely have something misconfigured or one of the dockers isnt staged properly. See if any of them have MFT of entries in the docker logs. I once had an issue where a docker wasnt able to start a program properly so it kept trying to restart it, which in turn ended up chewing through the log file like Gabe at a Buffet. For reference, my 3 dockers of Eggdrop, Nzbget, and Transmission are using all of 655 Megs after 1.5 months, going by: df -h /var/lib/docker
August 8, 201510 yr Author It is the Crashplan docker that is eating up the space. None of the others have this issue. Not sure what happened to this docker. At least I know the culprit now. david
August 8, 201510 yr My docker image is full but I can't figure out why: /dev/loop0 50G 49G 899M 99% /var/lib/docker But when I look at what is taking up the most space in /var/lib/docker I get this which doesn't add up: 13G containers 2.3G vfs 77M tmp 23M init 15M unraid 6.7M graph 1.5M volumes 32K linkgraph.db 8.0K repositories-btrfs 4.0K unraid-update-status.json So I'm not really sure what my issue is. Seems like there used to be a bug where the image filled up and space wasn't reclaimed but I'm having trouble searching for it now.
August 8, 201510 yr Author I updated to 6.1-rc2 as I saw docker got updated. I was on 6.0.3 Re-installing all containers and I'll see if crashplan still eats up the space. david
August 9, 201510 yr You need to use /data not /mnt for the docker side unless you edited the config as well.
August 9, 201510 yr Does this sound like the same issue? Not trying to hijack threads, but I came here with a very similar problem with a brand new Plex install from yesterday. Thanks! ug 9 06:57:09 Tower shfs/user0: shfs_mkdir: assign_disk: Applications/.plexmediaserver (28) No space left on device Aug 9 06:57:09 Tower logger: rsync: recv_generator: mkdir "/mnt/user0/Applications/.plexmediaserver" failed: No space left on device (28) Aug 9 06:57:09 Tower logger: *** Skipping any contents from this failed directory *** Aug 9 06:57:09 Tower logger: cd+++++++++ Applications/.plexmediaserver/ Aug 9 06:57:09 Tower logger: rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1165) [sender=3.1.0] Aug 9 06:57:09 Tower logger: ./Applications/.plexmediaserver/config/Plex Media Server/Media/localhost/4/076fe297b0b7106fc4b18da7f1b0325a97626df.bundle/Contents/Subtitle Contributions/com.plexapp.system Aug 9 06:57:09 Tower shfs/user0: shfs_mkdir: assign_disk: Applications/.plexmediaserver (28) No space left on device Aug 9 06:57:09 Tower logger: rsync: recv_generator: mkdir "/mnt/user0/Applications/.plexmediaserver" failed: No space left on device (28) Aug 9 06:57:09 Tower logger: *** Skipping any contents from this failed directory *** Aug 9 06:57:09 Tower logger: cd+++++++++ Applications/.plexmediaserver/ Aug 9 06:57:09 Tower logger: rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1165) [sender=3.1.0]
August 9, 201510 yr Does this sound like the same issue? Not trying to hijack threads, but I came here with a very similar problem with a brand new Plex install from yesterday. Thanks! ug 9 06:57:09 Tower shfs/user0: shfs_mkdir: assign_disk: Applications/.plexmediaserver (28) No space left on device Aug 9 06:57:09 Tower logger: rsync: recv_generator: mkdir "/mnt/user0/Applications/.plexmediaserver" failed: No space left on device (28) Aug 9 06:57:09 Tower logger: *** Skipping any contents from this failed directory *** Aug 9 06:57:09 Tower logger: cd+++++++++ Applications/.plexmediaserver/ Aug 9 06:57:09 Tower logger: rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1165) [sender=3.1.0] Aug 9 06:57:09 Tower logger: ./Applications/.plexmediaserver/config/Plex Media Server/Media/localhost/4/076fe297b0b7106fc4b18da7f1b0325a97626df.bundle/Contents/Subtitle Contributions/com.plexapp.system Aug 9 06:57:09 Tower shfs/user0: shfs_mkdir: assign_disk: Applications/.plexmediaserver (28) No space left on device Aug 9 06:57:09 Tower logger: rsync: recv_generator: mkdir "/mnt/user0/Applications/.plexmediaserver" failed: No space left on device (28) Aug 9 06:57:09 Tower logger: *** Skipping any contents from this failed directory *** Aug 9 06:57:09 Tower logger: cd+++++++++ Applications/.plexmediaserver/ Aug 9 06:57:09 Tower logger: rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1165) [sender=3.1.0] There is nothing in your Host paths that would result in this docker trying to access /mnt/user0/Applications/.plexmediaserver. Are you sure you don't have something else, maybe a Plex plugin still installed?
August 9, 201510 yr Author You need to use /data not /mnt for the docker side unless you edited the config as well. My config uses /mnt because I was using Crashplan before the docker was created. So when I moved to the docker, I just kept /mnt I did notice that the first volume was empty. Not sure how that happened, but I fixed that and am now running again. Not sure what happens when you have a <blank> mapped to <blank> david
August 9, 201510 yr Along the lines of what BRiT said, I have both of these volume mappings defined: /mnt -> /mnt /data -> /mnt I don't know if /data -> /mnt is needed, but I didn't delete it when adding /mnt -> /mnt. You could exec into the docker and see if a /data directory exists and has content. If so, that would confirm that the mapping is needed.
Archived
This topic is now archived and is closed to further replies.