Unexpected unclean shutdown

[facundomedica]

I have this error (which is shown several times) in dokku mongo:logs mydb:

e=true,path=journal,compressor=snappy),file_manager=(close_idle_time=100000),checkpoint=(wait=60,log_size=2GB),statistics_log=(wait=0),
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] MongoDB starting : pid=1 port=27017 dbpath=/data/db 64-bit host=b34bfc318ed3
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] db version v3.2.9
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] git version: 22ec9e93b40c85fc7cae7d56e7d6a02fd811088c
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] OpenSSL version: OpenSSL 1.0.1t  3 May 2016
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] allocator: tcmalloc
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] modules: none
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten] build environment:
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten]     distmod: debian81
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten]     distarch: x86_64
2017-02-17T15:02:06.533+0000 I CONTROL  [initandlisten]     target_arch: x86_64
2017-02-17T15:02:06.534+0000 I CONTROL  [initandlisten] options: { security: { authorization: "enabled" }, storage: { engine: "wiredTiger" } }
2017-02-17T15:02:06.558+0000 W -        [initandlisten] Detected unclean shutdown - /data/db/mongod.lock is not empty.
2017-02-17T15:02:06.559+0000 W STORAGE  [initandlisten] Recovering data from the last clean checkpoint.
2017-02-17T15:02:06.559+0000 I STORAGE  [initandlisten] wiredtiger_open config: create,cache_size=1G,session_max=20000,eviction=(threads_max=4),config_base=false,statistics=(fast),log=(enabled=true,archive=true,path=journal,compressor=snappy),file_manager=(close_idle_time=100000),checkpoint=(wait=60,log_size=2GB),statistics_log=(wait=0),

It just doesn't restart, and I can't run rm /data/db/mongod.lock because it says that is restarting. I didn't reboot my server or anything like that, I just deployed a new app and it happened.

[facundomedica]

Ok, after I deleted the previous mongo db and set up a new one, it happened again...

2017-02-20T12:59:49.245+0000 I CONTROL  [initandlisten] MongoDB starting : pid=1 port=27017 dbpath=/data/db 64-bit host=48d8502cafad
2017-02-20T12:59:49.250+0000 I CONTROL  [initandlisten] db version v3.2.9
2017-02-20T12:59:49.250+0000 I CONTROL  [initandlisten] git version: 22ec9e93b40c85fc7cae7d56e7d6a02fd811088c
2017-02-20T12:59:49.250+0000 I CONTROL  [initandlisten] OpenSSL version: OpenSSL 1.0.1t  3 May 2016
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten] allocator: tcmalloc
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten] modules: none
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten] build environment:
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten]     distmod: debian81
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten]     distarch: x86_64
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten]     target_arch: x86_64
2017-02-20T12:59:49.251+0000 I CONTROL  [initandlisten] options: { security: { authorization: "enabled" }, storage: { engine: "wiredTiger" } }
2017-02-20T12:59:49.331+0000 W -        [initandlisten] Detected unclean shutdown - /data/db/mongod.lock is not empty.
2017-02-20T12:59:49.331+0000 W STORAGE  [initandlisten] Recovering data from the last clean checkpoint.
2017-02-20T12:59:49.331+0000 I STORAGE  [initandlisten] wiredtiger_open config: create,cache_size=1G,session_max=20000,eviction=(threads_max=4),config_base=false,statistics=(fast),log=(enabled=true,archive=true,path=journal,compressor=snappy),file_manager=(close_idle_time=100000),checkpoint=(wait=60,log_size=2GB),statistics_log=(wait=0),

[josegonzalez]

Detected unclean shutdown - /data/db/mongod.lock is not empty.

Seems like you need to delete that file. How you got an unclean shutdown, I do not know. However, this really doesn't have anything to do with dokku, as honestly this is what we run:

docker run -e some=env_vars -v mounted:volumes mongo:YOUR_VERSION

[facundomedica]

Okay, I suspect that is maybe because of low RAM... (Somehow, if that makes any sense). If it happens again I'll try that and then try increasing my droplet's RAM. Thank you @josegonzalez !