Failed upgrade from 3.5.0.beta8-dev to 3.6.0.beta1

Installatie
1

I’m self-hosted and have 3 of the plugins that were recently moved to core.

My upgrade steps:

  1. Edited containers/app.yml and removed the lines for the 3 plugins (data-explorer, solved, assign)
  2. cd /var/discourse
  3. sudo ./launcher rebuild app

The rebuild takes the usual 5 minutes and seems to complete normally.

However, when I attempt to connect to the site, I get no response.

The last few lines in /var/discourse/shared/standalone/log/rails/production.log are:

Booted Rails 8.0.3 application in production environment
Running in ruby 3.3.8 (2025-04-09 revision b200bad6cd) [x86_64-linux]
See LICENSE and the LGPL-3.0 for licensing details.
Upgrade to Sidekiq Pro for more features and support: ``https://sidekiq.org
Sidekiq 7.3.9 connecting to Redis with options {:size=>5, :pool_name=>“default”, :host=>“localhost”, :port=>6379, :db=>1, :url=>nil}
Sidekiq 7.3.9 connecting to Redis with options {:size=>10, :pool_name=>“internal”, :host=>“localhost”, :port=>6379, :db=>1, :url=>nil}
Sidekiq 7.3.9 connecting to Redis with options {:size=>10, :pool_name=>“internal”, :host=>“localhost”, :port=>6379, :db=>1, :url=>nil}
Sidekiq 7.3.9 connecting to Redis with options {:size=>10, :pool_name=>“internal”, :host=>“localhost”, :port=>6379, :db=>1, :url=>nil}
Sidekiq 7.3.9 connecting to Redis with options {:size=>10, :pool_name=>“internal”, :host=>“localhost”, :port=>6379, :db=>1, :url=>nil}
Sidekiq 7.3.9 connecting to Redis with options {:size=>10, :pool_name=>“internal”, :host=>“localhost”, :port=>6379, :db=>1, :url=>nil}

The last few lines in /var/discourse/shared/standalone/log/rails/unicorn.stderr.log are:

/var/www/discourse/vendor/bundle/ruby/3.3.0/gems/message_bus-4.4.1/lib/message_bus/backends/redis.rb:291:in global_subscribe' /var/www/discourse/vendor/bundle/ruby/3.3.0/gems/message_bus-4.4.1/lib/message_bus.rb:769:in global_subscribe_thread’
/var/www/discourse/vendor/bundle/ruby/3.3.0/gems/message_bus-4.4.1/lib/message_bus.rb:740:in `block in new_subscriber_thread’
unknown OID 23281: failed to recognize type of ‘embeddings’. It will be treated as String.

I also tried rebuilding a 2nd time and its still not working.

Any idea what’s going on?

1 like
2

So the request times out?

3

The server refuses the connection. It seems to take about 2 seconds for the connection to be refused in both browsers I’ve tried. And nothing new shows up in production.log.

4

Any ideas what’s going on? Any suggestions for additional debugging steps?

5

I’d look at two things

  • is the server process running and listening
  • is your browser trying to contact the correct server

Check that your forum URL’s domain name does resolve, from your own PC, to the IP address of your server. Double-check.

Perhaps try, on your server

ping -c1 DOMAINNAME
netstat -l | egrep http
lsof -u root | egrep LISTEN
6

You can try

docker logs app

The last person I helped had a bogus value for his letsencrypt email address. It sounds like nginx isn’t starting up for some reason. The most common reason is that the certs are broken.