Rabbitmq: Is it possible to enable the plugins on startup via docker compose?

[GuyHarwood]

My compose file looks like this...

version: '3'
services:
  rabbit:
    image: "rabbitmq"
    container_name: rabbit
    ports:
    - "5672:5672"
    - "15672:15672"
    volumes:
    - "$PWD/rabbitmq/data/log:/data/log"
    - "$PWD/rabbitmq/data/mnesia:/data/mnesia"
# the following doesn't work, as it thinks its a directory....
    - "$PWD/rabbitmq/enabled_plugins:/etc/rabbitmq/enabled_plugins"

and i would like to enable the AMQP 1.0 protocol, which is one of the 'approved' plugins that (according to documentation) comes bundled with it.

Is there a way to enable this plugin by default (through argument, env var or conf file maybe?) when docker compose is started?

[yosifkit]

Does $PWD/rabbitmq/enabled_plugins exist before starting the container? Docker will create it as a directory if it does not exist on the host first.

You can also edit (generate) the file that contains the list of enabled plugins, in this case /etc/rabbitmq/enabled_plugins. It's an Erlang list of atoms followed by a dot. There is nothing wrong with modifying that file during deployment before the node is started.

- https://github.com/docker-library/rabbitmq/issues/145#issuecomment-291476039

[GuyHarwood]

Yes, it does, that's the odd thing...

/dev/spike-rabbit/rabbitmq> ls                                                                        
total 32
drwxr-xr-x  5  staff   160B 23 May 17:14 ./
drwxr-xr-x  7  staff   224B 23 May 17:08 ../
-rw-r--r--@ 1 staff   8.0K 23 May 17:23 .DS_Store
drwxr-xr-x  5  staff   160B 23 May 17:14 data/
-rw-r--r--  1  staff    72B 23 May 17:17 enabled_plugins

the content of enabled_plugins....

[rabbitmq_management,rabbitmq_management_visualiser, rabbitmq_amqp1_0].

and the runtime error....

> docker-compose up
Starting rabbit ... error
ERROR: for rabbit  Cannot start service rabbit: OCI runtime create failed: 
container_linux.go:348: starting container process caused "process_linux.go:402: 
container init caused \"rootfs_linux.go:58: 
mounting \\\"//dev/mtc/spike-rabbit/rabbitmq/enabled_plugins\\\" to rootfs 
\\\"/var/lib/docker/overlay2/49295881d7680ecb1f7d289666d84c35cf05e5476276ad1a286bfdc7bc6e7374/merged\\\" at 
\\\"/var/lib/docker/overlay2/49295881d7680ecb1f7d289666d84c35cf05e5476276ad1a286bfdc7bc6e7374/merged/etc/rabbitmq/enabled_plugins\\\" caused \\\"not a directory\\\"\"":
 unknown: Are you trying to mount a directory onto a file (or vice-versa)? 
Check if the specified host path exists and is the expected type

[yosifkit]

Are you certain that you are using the image from this repo? The Dockerfile does not use anything from /data, but instead uses a volume at /var/lib/rabbitmq. Have you checked inside the image to see if enabled_plugins is a file (or non-existent) and not a folder?

$ docker pull rabbitmq:3.7.5
3.7.5: Pulling from library/rabbitmq
f2aa67a397c4: Pull complete 
f062288ad968: Pull complete 
8b9469379b84: Pull complete 
5b66af38c756: Pull complete 
d942356ed811: Pull complete 
dd730b88925c: Pull complete 
353d0c7f2496: Pull complete 
8f664faefe5e: Pull complete 
e52f50b7979d: Pull complete 
c785d5ce338c: Pull complete 
be6d0e07a9f4: Pull complete 
31cf1b609ffd: Pull complete 
Digest: sha256:e9c89a727063acfdbcf885e22f0035072dc1faac2b7737da187af3685b24ea52
Status: Downloaded newer image for rabbitmq:3.7.5
$ docker run -it --rm rabbitmq:3.7.5 ls -al /etc/rabbitmq/
total 8
drwxrwsrwx 1 rabbitmq rabbitmq 4096 May 10 03:05 .
drwxr-xr-x 1 root     root     4096 May 24 21:09 ..

I am unable to reproduce the observed failure:

$ echo '[rabbitmq_management,rabbitmq_management_visualiser, rabbitmq_amqp1_0].' > enabled_plugins
$ docker run -it --rm -v "$PWD/enabled_plugins:/etc/rabbitmq/enabled_plugins" rabbitmq:3.7.5
2018-05-24 21:03:24.836 [info] <0.33.0> Application lager started on node rabbit@3b6637744abf
2018-05-24 21:03:24.967 [info] <0.33.0> Application xmerl started on node rabbit@3b6637744abf
2018-05-24 21:03:24.967 [info] <0.33.0> Application recon started on node rabbit@3b6637744abf
2018-05-24 21:03:25.022 [info] <0.33.0> Application mnesia started on node rabbit@3b6637744abf
2018-05-24 21:03:25.057 [info] <0.33.0> Application inets started on node rabbit@3b6637744abf
2018-05-24 21:03:25.057 [info] <0.33.0> Application crypto started on node rabbit@3b6637744abf
2018-05-24 21:03:25.057 [info] <0.33.0> Application amqp10_common started on node rabbit@3b6637744abf
2018-05-24 21:03:25.057 [info] <0.33.0> Application jsx started on node rabbit@3b6637744abf
2018-05-24 21:03:25.059 [info] <0.33.0> Application os_mon started on node rabbit@3b6637744abf
2018-05-24 21:03:25.059 [info] <0.33.0> Application cowlib started on node rabbit@3b6637744abf
2018-05-24 21:03:25.059 [info] <0.33.0> Application asn1 started on node rabbit@3b6637744abf
2018-05-24 21:03:25.060 [info] <0.33.0> Application public_key started on node rabbit@3b6637744abf
2018-05-24 21:03:25.089 [info] <0.33.0> Application ssl started on node rabbit@3b6637744abf
2018-05-24 21:03:25.091 [info] <0.33.0> Application ranch started on node rabbit@3b6637744abf
2018-05-24 21:03:25.092 [info] <0.33.0> Application ranch_proxy_protocol started on node rabbit@3b6637744abf
2018-05-24 21:03:25.093 [info] <0.33.0> Application cowboy started on node rabbit@3b6637744abf
2018-05-24 21:03:25.093 [info] <0.33.0> Application rabbit_common started on node rabbit@3b6637744abf
2018-05-24 21:03:25.098 [info] <0.33.0> Application amqp_client started on node rabbit@3b6637744abf
2018-05-24 21:03:25.104 [info] <0.194.0> 
 Starting RabbitMQ 3.7.5 on Erlang 20.3.5
 Copyright (C) 2007-2018 Pivotal Software, Inc.
 Licensed under the MPL.  See http://www.rabbitmq.com/

  ##  ##
  ##  ##      RabbitMQ 3.7.5. Copyright (C) 2007-2018 Pivotal Software, Inc.
  ##########  Licensed under the MPL.  See http://www.rabbitmq.com/
  ######  ##
  ##########  Logs: <stdout>

              Starting broker...
2018-05-24 21:03:25.113 [info] <0.194.0> 
 node           : rabbit@3b6637744abf
 home dir       : /var/lib/rabbitmq
 config file(s) : /etc/rabbitmq/rabbitmq.conf
 cookie hash    : ZuULuWqEPwfwaqn+OeDEug==
 log(s)         : <stdout>
 database dir   : /var/lib/rabbitmq/mnesia/rabbit@3b6637744abf
2018-05-24 21:03:26.224 [info] <0.202.0> Memory high watermark set to 12643 MiB (13257608396 bytes) of 31608 MiB (33144020992 bytes) total
2018-05-24 21:03:26.226 [info] <0.204.0> Enabling free disk space monitoring
2018-05-24 21:03:26.226 [info] <0.204.0> Disk free limit set to 50MB
2018-05-24 21:03:26.227 [info] <0.206.0> Limiting to approx 1048476 file handles (943626 sockets)
2018-05-24 21:03:26.228 [info] <0.207.0> FHC read buffering:  OFF
2018-05-24 21:03:26.228 [info] <0.207.0> FHC write buffering: ON
2018-05-24 21:03:26.228 [info] <0.194.0> Node database directory at /var/lib/rabbitmq/mnesia/rabbit@3b6637744abf is empty. Assuming we need to join an existing cluster or initialise from scratch...
2018-05-24 21:03:26.228 [info] <0.194.0> Configured peer discovery backend: rabbit_peer_discovery_classic_config
2018-05-24 21:03:26.228 [info] <0.194.0> Will try to lock with peer discovery backend rabbit_peer_discovery_classic_config
2018-05-24 21:03:26.228 [info] <0.194.0> Peer discovery backend does not support locking, falling back to randomized delay
2018-05-24 21:03:26.229 [info] <0.194.0> Peer discovery backend rabbit_peer_discovery_classic_config does not support registration, skipping randomized startup delay.
2018-05-24 21:03:26.229 [info] <0.194.0> All discovered existing cluster peers: 
2018-05-24 21:03:26.229 [info] <0.194.0> Discovered no peer nodes to cluster with
2018-05-24 21:03:26.230 [info] <0.33.0> Application mnesia exited with reason: stopped
2018-05-24 21:03:26.237 [info] <0.33.0> Application mnesia started on node rabbit@3b6637744abf
2018-05-24 21:03:26.273 [info] <0.194.0> Waiting for Mnesia tables for 30000 ms, 9 retries left
2018-05-24 21:03:26.307 [info] <0.194.0> Waiting for Mnesia tables for 30000 ms, 9 retries left
2018-05-24 21:03:26.328 [info] <0.194.0> Waiting for Mnesia tables for 30000 ms, 9 retries left
2018-05-24 21:03:26.328 [info] <0.194.0> Peer discovery backend rabbit_peer_discovery_classic_config does not support registration, skipping registration.
2018-05-24 21:03:26.328 [info] <0.194.0> Priority queues enabled, real BQ is rabbit_variable_queue
2018-05-24 21:03:26.330 [info] <0.381.0> Starting rabbit_node_monitor
2018-05-24 21:03:26.349 [info] <0.194.0> message_store upgrades: 1 to apply
2018-05-24 21:03:26.350 [info] <0.194.0> message_store upgrades: Applying rabbit_variable_queue:move_messages_to_vhost_store
2018-05-24 21:03:26.350 [info] <0.194.0> message_store upgrades: No durable queues found. Skipping message store migration
2018-05-24 21:03:26.350 [info] <0.194.0> message_store upgrades: Removing the old message store data
2018-05-24 21:03:26.350 [info] <0.194.0> message_store upgrades: All upgrades applied successfully
2018-05-24 21:03:26.370 [info] <0.194.0> Management plugin: using rates mode 'basic'
2018-05-24 21:03:26.371 [info] <0.194.0> Adding vhost '/'
2018-05-24 21:03:26.378 [info] <0.417.0> Making sure data directory '/var/lib/rabbitmq/mnesia/rabbit@3b6637744abf/msg_stores/vhosts/628WB79CIFDYO9LJI6DKMI09L' for vhost '/' exists
2018-05-24 21:03:26.379 [info] <0.417.0> Starting message stores for vhost '/'
2018-05-24 21:03:26.379 [info] <0.421.0> Message store "628WB79CIFDYO9LJI6DKMI09L/msg_store_transient": using rabbit_msg_store_ets_index to provide index
2018-05-24 21:03:26.380 [info] <0.417.0> Started message store of type transient for vhost '/'
2018-05-24 21:03:26.380 [info] <0.424.0> Message store "628WB79CIFDYO9LJI6DKMI09L/msg_store_persistent": using rabbit_msg_store_ets_index to provide index
2018-05-24 21:03:26.380 [warning] <0.424.0> Message store "628WB79CIFDYO9LJI6DKMI09L/msg_store_persistent": rebuilding indices from scratch
2018-05-24 21:03:26.381 [info] <0.417.0> Started message store of type persistent for vhost '/'
2018-05-24 21:03:26.381 [info] <0.194.0> Creating user 'guest'
2018-05-24 21:03:26.383 [info] <0.194.0> Setting user tags for user 'guest' to [administrator]
2018-05-24 21:03:26.384 [info] <0.194.0> Setting permissions for 'guest' in '/' to '.*', '.*', '.*'
2018-05-24 21:03:26.385 [info] <0.462.0> started TCP Listener on [::]:5672
2018-05-24 21:03:26.387 [info] <0.194.0> Setting up a table for connection tracking on this node: tracked_connection_on_node_rabbit@3b6637744abf
2018-05-24 21:03:26.388 [info] <0.194.0> Setting up a table for per-vhost connection counting on this node: tracked_connection_per_vhost_on_node_rabbit@3b6637744abf
2018-05-24 21:03:26.389 [info] <0.33.0> Application rabbit started on node rabbit@3b6637744abf
2018-05-24 21:03:26.389 [info] <0.33.0> Application rabbitmq_amqp1_0 started on node rabbit@3b6637744abf
2018-05-24 21:03:26.390 [info] <0.33.0> Application rabbitmq_management_agent started on node rabbit@3b6637744abf
2018-05-24 21:03:26.390 [info] <0.33.0> Application rabbitmq_web_dispatch started on node rabbit@3b6637744abf
2018-05-24 21:03:26.409 [info] <0.520.0> Management plugin started. Port: 15672
2018-05-24 21:03:26.409 [info] <0.626.0> Statistics database started.
2018-05-24 21:03:26.409 [info] <0.33.0> Application rabbitmq_management started on node rabbit@3b6637744abf
 completed with 4 plugins.
2018-05-24 21:03:26.496 [info] <0.5.0> Server startup complete; 4 plugins started.
 * rabbitmq_management
 * rabbitmq_web_dispatch
 * rabbitmq_management_agent
 * rabbitmq_amqp1_0
^CStopping and halting node rabbit@3b6637744abf ...

[GuyHarwood]

bingo.

i translated what you did into my compose file, and it's working.

The only real difference being that the enabled_plugins file is at the root with the compose file, rather than 1 directory deep. If i push it 1 directory deep, it fails again, but when in same location as compose file, it works.

Thanks for your help with this, it's appreciated.

[yosifkit]

Not certain how we fixed it (since host file location shouldn't affect this), but glad it is working.

I moved up a directory and it still worked fine.

$ docker run -it --rm -v "$PWD/rabbitmq/enabled_plugins:/etc/rabbitmq/enabled_plugins" rabbitmq:3.7.5

[thiagoufg]

bingo.

i translated what you did into my compose file, and it's working.

The only real difference being that the enabled_plugins file is at the root with the compose file, rather than 1 directory deep. If i push it 1 directory deep, it fails again, but when in same location as compose file, it works.

Thanks for your help with this, it's appreciated.

Can you share your translated docker-compose file?

[GuyHarwood]

@thiagoufg sure...

version: '3'
services:
  sqldb:
    image: "microsoft/mssql-server-linux:2017-GDR"
    container_name: mtc_mssql
    ports:
    - "1433:1433"
    environment:
        SA_PASSWORD: "mypassword"
        ACCEPT_EULA: "Y"
        HOMEBREW_NO_ENV_FILTERING: 1
  rabbit:
    image: "rabbitmq"
    container_name: rabbit
    ports:
    - "5672:5672"
    - "15672:15672"
    volumes:
    - "$PWD/rabbitmq/data/log:/data/log"
    - "$PWD/rabbitmq/data/mnesia:/data/mnesia"
    - "$PWD/rabbit_enabled_plugins:/etc/rabbitmq/enabled_plugins"

[nsarvar]

  FROM rabbitmq:3.7.18-management
  COPY ./rabbitmq_delayed_message_exchange-20171201-3.7.x.ez /opt/rabbitmq/plugins/
  RUN rabbitmq-plugins enable rabbitmq_delayed_message_exchange