Docker container issues on any update past Unraid 6.7.2

[benrenewed]

I am currently running Unraid 6.7.2 and am eager to update to the latest version in order to get some of the new features. But every time I update past 6.7.2, I run into issues with one of my Docker containers. It is an image coming from the Docker Hub, so there's not a corresponding support thread available for it. But neither the Docker image nor the application it contains are receiving any updates. See Docker container command below. Docker Hub link: https://hub.docker.com/r/fflo/mfi-controller/

root@localhost:# /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker run -d --name='mfi-controller-previous' --net='bridge' --privileged=true -e TZ="America/Chicago" -e HOST_OS="Unraid" -p '6080:6080/tcp' -p '6443:6443/tcp' -p '3478:3478/udp' -p '2323:2323/tcp' -p '6843:6843/tcp' -p '6880:6880/tcp' -p '10001:10001/udp' -v '/mnt/user/appdata/mfi':'/var/lib/mfi':'rw' 'fflo/mfi-controller'

bd03125da4fda3da9c0fc02637fb61189edf7bbe9a3221a4c2b46106acf070cb

As noted above, updating Unraid to any version past 6.7.2 causes issues with this container. Rolling Unraid back to 6.7.2 and restarting restores functionality to the container.

 

I compared the Docker container logs between the old and updated Unraid versions and found the below. The application in the container uses MongoDB to store its data, and it appears that it runs into an issue when starting mongod. Fatal Assertion 13515 appears to be related to the disk being full (or perhaps when a write fails?). As it works fine on the older Unraid version, I suspect that the issue may be related to the interaction between the container and Docker or some other process in Unraid.

I tried to broaden the file permissions of the /mnt/user/appdata/mfi directory to no avail.

[2020-04-25 09:19:53,351] <DbServer> ERROR B - [exec] error, rc=14, output=all output going to: /usr/lib/mfi/logs/mongod.log
***** SERVER RESTARTED *****

Sat Apr 25 09:19:53.338 [initandlisten] MongoDB starting : pid=37 port=37017 dbpath=/usr/lib/mfi/data/db 64-bit host=unraid
Sat Apr 25 09:19:53.338 [initandlisten] db version v2.4.10Sat Apr 25 09:19:53.338 [initandlisten] git version: nogitversion
Sat Apr 25 09:19:53.338 [initandlisten] build info: Linux x86-ubc-01 3.16.0-4-amd64 #1 SMP Debian 3.16.39-1 (2016-12-30) x86_64 BOOST_LIB_VERSION=1_55
Sat Apr 25 09:19:53.338 [initandlisten] allocator: tcmalloc
Sat Apr 25 09:19:53.338 [initandlisten] options: { bind_ip: "127.0.0.1", dbpath: "/usr/lib/mfi/data/db", logappend: true, logpath: "logs/mongod.log", nohttpinterface: true, port: 37017 }
Sat Apr 25 09:19:53.345 [initandlisten] journal dir=/usr/lib/mfi/data/db/journal
Sat Apr 25 09:19:53.345 [initandlisten] recover : no journal files present, no recovery needed
Sat Apr 25 09:19:53.347 [initandlisten] LogFile::synchronousAppend failed with 8192 bytes unwritten out of 8192 bytes; b=0x36ce000 errno:14 Bad address
Sat Apr 25 09:19:53.347 [initandlisten] Fatal Assertion 13515
            0xa6bcc6        0xa3e63d        0xa2230d        0xa42a22        0x7282b2        0x72871b        0x72d7cf        0x721b50        0x5866c7        0x588dff        0x5664f1        0x1498f27f6b45  0x57e05c
            bin/mongod(_ZN5mongo15printStackTraceERSo+0x26)         [0xa6bcc6]
            bin/mongod(_ZN5mongo10logContextEPKc+0xfd)             [0xa3e63d]
            bin/mongod(_ZN5mongo13fassertFailedEi+0x4d)             [0xa2230d]
            bin/mongod(_ZN5mongo7LogFile17synchronousAppendEPKvm+0x1c2)     [0xa42a22]
            bin/mongod(_ZN5mongo3dur20_preallocateIsFasterEv+0x102)     [0x7282b2]
            bin/mongod(_ZN5mongo3dur19preallocateIsFasterEv+0x1b)         [0x72871b]
            bin/mongod(_ZN5mongo3dur16preallocateFilesEv+0x3ef)         [0x72d7cf]
            bin/mongod(_ZN5mongo3dur7startupEv+0x30)             [0x721b50]
            bin/mongod(_ZN5mongo14_initAndListenEi+0x8c7)             [0x5866c7]
            bin/mongod(_ZN5mongo13initAndListenEi+0xf)             [0x588dff]
            bin/mongod(main+0x2f1)                         [0x5664f1]
            /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5)     [0x1498f27f6b45]
            bin/mongod()                             [0x57e05c]
Sat Apr 25 09:19:53.348 [initandlisten] ***aborting after fassert() failure
Sat Apr 25 09:19:53.348 Got signal: 6 (Aborted).
Sat Apr 25 09:19:53.350 Backtrace:
            0xa6bcc6        0x57ecda        0x1498f280a0e0  0x1498f280a067  0x1498f280b448  0xa22348        0xa42a22        0x7282b2        0x72871b        0x72d7cf        0x721b50        0x5866c7        0x588dff        0x5664f1        0x1498f27f6b45  0x57e05c
            bin/mongod(_ZN5mongo15printStackTraceERSo+0x26)         [0xa6bcc6]
            bin/mongod(_ZN5mongo10abruptQuitEi+0x36a)             [0x57ecda]
            /lib/x86_64-linux-gnu/libc.so.6(+0x350e0)             [0x1498f280a0e0]
            /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x37)             [0x1498f280a067]
            /lib/x86_64-linux-gnu/libc.so.6(abort+0x148)             [0x1498f280b448]
            bin/mongod(_ZN5mongo13fassertFailedEi+0x88)             [0xa22348]
            bin/mongod(_ZN5mongo7LogFile17synchronousAppendEPKvm+0x1c2)     [0xa42a22]
            bin/mongod(_ZN5mongo3dur20_preallocateIsFasterEv+0x102)     [0x7282b2]
            bin/mongod(_ZN5mongo3dur19preallocateIsFasterEv+0x1b)         [0x72871b]
            bin/mongod(_ZN5mongo3dur16preallocateFilesEv+0x3ef)         [0x72d7cf]
            bin/mongod(_ZN5mongo3dur7startupEv+0x30)             [0x721b50]
            bin/mongod(_ZN5mongo14_initAndListenEi+0x8c7)             [0x5866c7]
            bin/mongod(_ZN5mongo13initAndListenEi+0xf)             [0x588dff]
            bin/mongod(main+0x2f1)                         [0x5664f1]
            /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5)     [0x1498f27f6b45]
            bin/mongod()                             [0x57e05c]

I would appreciate any tips for things to look into or any common "gotcha" scenarios. If more information or log dumps, etc. are necessary just ask. I apologize if this thread was posted in the wrong category, or the wrong forum altogether. I'm simply looking for suggestions/answers and appreciate your time.

[Squid]

Try changing /mnt/user/appdata/mfi in the template settings to instead be /mnt/cache/appdata/mfi (or if you don't have a cache drive, then /mnt/diskX/appdata/mfi'

[benrenewed]

Bless you Squid, that solved it!

I think I've seen you mention the /mnt/user/appdata/xxx vs /mnt/cache/appdata/xxx path before. Not sure how that creates a conflict, but that's okay.

For what it's worth, I do have a single ssd as a cache drive and the appdata share is set to "Prefer" cache so it'll (theoretically) have somewhere to spill over to.

(That's not a configuration question per se, but more information for anyone else who sees this thread later.)

[Squid]

23 minutes ago, benrenewed said:

Not sure how that creates a conflict,

If somebody somewhere could possibly answer why this is needed for a subset of users, then it could get fixed