Docker Compose Redis exits immediately

[SiqingYu]

Here is the output of docker-compose up:

user@user-XPS-13-9350:~/repos/NewsBlur$ docker-compose up
Starting newsblur_mongo_1         ... done
Starting newsblur_elasticsearch_1 ... done
Starting newsblur_postgres_1      ... done
Recreating newsblur_redis_1       ... done
Recreating newsblur_newsblur_1    ... done
Attaching to newsblur_postgres_1, newsblur_mongo_1, newsblur_elasticsearch_1, newsblur_redis_1, newsblur_newsblur_1
postgres_1       | LOG:  database system was shut down at 2018-11-04 15:39:47 UTC
postgres_1       | LOG:  MultiXact member wraparound protections are now enabled
postgres_1       | LOG:  database system is ready to accept connections
postgres_1       | LOG:  autovacuum launcher started
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] MongoDB starting : pid=1 port=27017 dbpath=/data/db 64-bit host=5ff0ca4d1710
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] db version v3.2.21
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] git version: 1ab1010737145ba3761318508ff65ba74dfe8155
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] OpenSSL version: OpenSSL 1.0.1t  3 May 2016
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] allocator: tcmalloc
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] modules: none
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] build environment:
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten]     distmod: debian81
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten]     distarch: x86_64
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten]     target_arch: x86_64
mongo_1          | 2018-11-04T15:43:14.778+0000 I CONTROL  [initandlisten] options: { storage: { mmapv1: { smallFiles: true } } }
mongo_1          | 2018-11-04T15:43:14.782+0000 I -        [initandlisten] Detected data files in /data/db created by the 'wiredTiger' storage engine, so setting the active storage engine to 'wiredTiger'.
mongo_1          | 2018-11-04T15:43:14.783+0000 I STORAGE  [initandlisten] wiredtiger_open config: create,cache_size=8G,session_max=20000,eviction=(threads_min=4,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),verbose=(recovery_progress),
mongo_1          | 2018-11-04T15:43:14.922+0000 I STORAGE  [initandlisten] WiredTiger [1541346194:922284][1:0x7fe8d8c7acc0], txn-recover: Main recovery loop: starting at 6/4224
mongo_1          | 2018-11-04T15:43:14.996+0000 I STORAGE  [initandlisten] WiredTiger [1541346194:996871][1:0x7fe8d8c7acc0], txn-recover: Recovering log 6 through 7
mongo_1          | 2018-11-04T15:43:15.000+0000 I STORAGE  [initandlisten] WiredTiger [1541346195:775][1:0x7fe8d8c7acc0], txn-recover: Recovering log 7 through 7
mongo_1          | 2018-11-04T15:43:15.224+0000 W STORAGE  [initandlisten] Detected configuration for non-active storage engine mmapv1 when current storage engine is wiredTiger
mongo_1          | 2018-11-04T15:43:15.226+0000 I FTDC     [initandlisten] Initializing full-time diagnostic data capture with directory '/data/db/diagnostic.data'
mongo_1          | 2018-11-04T15:43:15.226+0000 I NETWORK  [HostnameCanonicalizationWorker] Starting hostname canonicalization worker
mongo_1          | 2018-11-04T15:43:15.226+0000 I NETWORK  [initandlisten] waiting for connections on port 27017
elasticsearch_1  | [2018-11-04 15:43:15,365][INFO ][node                     ] [Ramshot] version[1.7.6], pid[1], build[c730b59/2016-11-18T15:21:16Z]
elasticsearch_1  | [2018-11-04 15:43:15,366][INFO ][node                     ] [Ramshot] initializing ...
elasticsearch_1  | [2018-11-04 15:43:15,421][INFO ][plugins                  ] [Ramshot] loaded [], sites []
elasticsearch_1  | [2018-11-04 15:43:15,450][INFO ][env                      ] [Ramshot] using [1] data paths, mounts [[/usr/share/elasticsearch/data (/dev/nvme0n1p2)]], net usable_space [88.8gb], net total_space [233.2gb], types [ext4]
newsblur_redis_1 exited with code 0
newsblur_1       | [2018-11-04 15:43:16 +0000] [1] [INFO] Starting gunicorn 19.7.0
newsblur_1       | [2018-11-04 15:43:16 +0000] [1] [INFO] Listening at: http://0.0.0.0:8000 (1)
newsblur_1       | [2018-11-04 15:43:16 +0000] [1] [INFO] Using worker: sync
newsblur_1       | [2018-11-04 15:43:16 +0000] [11] [INFO] Booting worker with pid: 11
elasticsearch_1  | [2018-11-04 15:43:17,441][INFO ][node                     ] [Ramshot] initialized
elasticsearch_1  | [2018-11-04 15:43:17,442][INFO ][node                     ] [Ramshot] starting ...
elasticsearch_1  | [2018-11-04 15:43:17,539][INFO ][transport                ] [Ramshot] bound_address {inet[/0.0.0.0:9300]}, publish_address {inet[/172.20.0.4:9300]}
elasticsearch_1  | [2018-11-04 15:43:17,567][INFO ][discovery                ] [Ramshot] elasticsearch/RBtSRUJ2T1GxhnSAMH9udg
elasticsearch_1  | [2018-11-04 15:43:21,367][INFO ][cluster.service          ] [Ramshot] new_master [Ramshot][RBtSRUJ2T1GxhnSAMH9udg][47dec3105eab][inet[/172.20.0.4:9300]], reason: zen-disco-join (elected_as_master)
elasticsearch_1  | [2018-11-04 15:43:21,417][INFO ][http                     ] [Ramshot] bound_address {inet[/0.0.0.0:9200]}, publish_address {inet[/172.20.0.4:9200]}
elasticsearch_1  | [2018-11-04 15:43:21,418][INFO ][node                     ] [Ramshot] started
elasticsearch_1  | [2018-11-04 15:43:21,439][INFO ][gateway                  ] [Ramshot] recovered [0] indices into cluster_state
newsblur_1       | [2018-11-04 15:54:50 +0000] [1] [CRITICAL] WORKER TIMEOUT (pid:11)
newsblur_1       | [2018-11-04 15:54:50 +0000] [11] [INFO] Worker exiting (pid: 11)
newsblur_1       | [2018-11-04 15:54:51 +0000] [12] [INFO] Booting worker with pid: 12

The Redis container does not seem to leave any logs of the exit:

user@user-XPS-13-9350:~/repos/NewsBlur$ docker-compose logs --follow redis
Attaching to newsblur_redis_1
newsblur_redis_1 exited with code 0

[samuelclay]

Thanks @leophys! Any chance you can take a look at why the postgres user isn't being created? See https://forum.newsblur.com/t/setting-up-newsblur-on-docker/6919

[leophys]

Thanks @leophys! Any chance you can take a look at why the postgres user isn't being created? See https://forum.newsblur.com/t/setting-up-newsblur-on-docker/6919

Dear @samuelclay

The sql dump with which the db is initialized has two statements with the user postgres hardcoded at the end of the file (lines 4144-4145):

REVOKE ALL ON SCHEMA public FROM postgres;
GRANT ALL ON SCHEMA public TO postgres;

But the container is built (in the docker-compose.yml file) using another user to launch and manage the db (POSTGRES_USER=newsblur). The error one gets is:

postgres_1       | ERROR:  role "postgres" does not exist                                                                                                                                                          
postgres_1       | STATEMENT:  REVOKE ALL ON SCHEMA public FROM postgres;                                                                                                                                        
postgres_1       | ERROR:  role "postgres" does not exist  

The solution would be either to change the user in the sql dump from postgres to newsblur or to remove the above POSTGRES_USER environment variable, therefore managing the database with the standard user postgres in the container. If you use directly this newsblur user around, I suggest the first solution. I did it but the file is shipped gzipped and therefore the diff would not be shown in the commit diff. If you are comfortable with this I can arrange another PR. Otherwise proceed autonomously with one of these changes 😊

[samuelclay]

Sure, go ahead and send that as a PR. I'll unzip it myself and double-check, but I love having more contributors.