Opened 7 years ago
Closed 4 years ago
#5520 closed defect (no-action)
celery has wrong broker in config files
Reported by: | Bruno Santos | Owned by: | |
---|---|---|---|
Priority: | major | Milestone: | |
Component: | programming | Keywords: | |
Cc: | Parent Tickets: |
Description
After installing mediagoblin and used the systemd init files,
I wasn't able to transcode any files..
After checking the logs /var/log/messages
I get the following error:
ERROR/MainProcess] consumer: Cannot connect to amqp://guest:**@127.0.0.1:5672//: [Errno 111] Connection refused. Trying again in 14.00 seconds... [2017-07-28 17:39:06,874: ERROR/MainProcess] consumer: Cannot connect to amqp://guest:**@127.0.0.1:5672//: [Errno 111] Connection refused. Trying again in 16.00 seconds... [2017-07-28 17:39:22,912: ERROR/MainProcess] consumer: Cannot connect to amqp://guest:**@127.0.0.1:5672//: [Errno 111] Connection refused. Trying again in 18.00 seconds.
After searching through the issues, I found issue #859:
https://issues.mediagoblin.org/ticket/859
somewhat related to this problem...
I've checked config_spec.ini and found this:
[celery] # default result stuff CELERY_RESULT_BACKEND = string(default="database") CELERY_RESULT_DBURI = string(default="sqlite:///%(here)s/celery.db") # default kombu stuff BROKER_URL = string(default="amqp://")
After changing the kombu stuff to:
BROKER_TRANSPORT = string(default="sqlalchemy") BROKER_URL = string(default="sqlite:///%(here)s/kombu.db")
now it works...
I've never touched this file, this is configuration by default and there's nothing in the easy deployment wiki page to change something in here...
Change History (2)
comment:1 by , 7 years ago
comment:2 by , 4 years ago
Resolution: | → no-action-required |
---|---|
Status: | new → closed |
Thanks Bruno and Elinvention and I'm sorry you didn't get a reply to this issue at the time. That's right that RabbitMQ is now the default broker on new installations.
Closing this as we haven't had any recent reports of installation problems.
I had the same problem, then I found out that devs made the decision to switch to RabbitMQ, instead of using a sqlite database as a queue for celery's tasks. The discussion is here: 2016-09-04 meeting and 2017-03-04 meeting.
So those lines are needed if you want to keep the deprecated sqlite middleware, or you can
and that's it.