A user, silverdr, is trying to set up Discourse with a separate Redis server on a local network, having already configured PostgreSQL successfully. Despite following instructions to use an external Redis by setting DISCOURSE_REDIS_HOST, the setup fails during the db:migrate step with a Pups::ExecError. The rebuild log shows that the issue stems from Redis version compatibility—Discourse requires Redis 6.2.0 or higher, but the system’s Ubuntu 22.04.2 LTS provides version 6.0.16. Falco suggests checking for password, TLS, or connectivity issues, while renato confirms the version requirement and recommends installing Redis 6.2+ via Redis’ official APT repository. silverdr notes that the error message is misleading, appearing as an INFO log rather than a critical failure, which caused confusion. pfaffman advises using the built-in Redis to avoid complexity, but the user is focused on optimizing resource usage by leveraging the existing database server.
There is a half-lazy database server running PostgreSQL inside local network and I wanted to use it for Discourse too. Reason for this is that the web server, is far more stressed than the previously mentioned database one, which still has a lot of unused resources/potential. Redirecting Discourse’s PostgreSQL was quite easy, following a few threads on this forum (example). Whenever I try to do the same for Redis though:
commenting out - "templates/redis.template.yml"
setting DISCOURSE_REDIS_HOST instead
I always get:
Pups::ExecError: cd /var/www/discourse && su discourse -c 'bundle exec rake db:migrate' failed with return #<Process::Status: pid 582 exit 1>
Location of failure: /usr/local/lib/ruby/gems/3.2.0/gems/pups-1.1.1/lib/pups/exec_command.rb:117:in `spawn'
exec failed with the params {"cd"=>"$home", "hook"=>"db_migrate", "cmd"=>["su discourse -c 'bundle exec rake db:migrate'"]}
bootstrap failed with exit code 1
whenever I try to rebuild with launcher rebuild app
Is this a known issue? Anyone successfully running Discourse with non-containerised Redis?
That is all you need indeed. Can you paste a the complete rebuild error? You probably have either the need for a password, mandatory TLS or an unreachable domain. Which one should be clear in the rebuild log.
That’s basically all. There are some warnings further up the log:
warning " > @mixer/parallel-prettier@2.0.3" has unmet peer dependency "prettier@^2.0.0".
warning "eslint-config-discourse > eslint-plugin-lodash@7.1.0" has unmet peer dependency "lodash@>=4".
warning " > babel-plugin-debug-macros@0.4.0-pre1" has unmet peer dependency "@babel/core@^7.0.0".
warning "workspace-aggregator-3fff36d0-9fd8-4d1a-ac4c-d10694a59627 > discourse > @uppy/aws-s3@3.0.6" has incorrect peer dependency "@uppy/core@^3.1.2".
warning "workspace-aggregator-3fff36d0-9fd8-4d1a-ac4c-d10694a59627 > discourse > @uppy/aws-s3-multipart@3.1.3" has incorrect peer dependency "@uppy/core@^3.1.2".
warning "workspace-aggregator-3fff36d0-9fd8-4d1a-ac4c-d10694a59627 > discourse > @uppy/xhr-upload@3.1.1" has incorrect peer dependency "@uppy/core@^3.1.2".
but they are there even when rebuilding with containerised Redis.
FAILED
--------------------
Pups::ExecError: cd /var/www/discourse && su discourse -c 'bundle exec rake db:migrate' failed with return #<Process::Status: pid 581 exit 1>
Location of failure: /usr/local/lib/ruby/gems/3.2.0/gems/pups-1.1.1/lib/pups/exec_command.rb:117:in `spawn'
exec failed with the params {"cd"=>"$home", "hook"=>"db_migrate", "cmd"=>["su discourse -c 'bundle exec rake db:migrate'"]}
bootstrap failed with exit code 1
** FAILED TO BOOTSTRAP ** please scroll up and look for earlier error messages, there may be more than one.
./discourse-doctor may help diagnose the problem.
2533af220cc44d8c77bc12c17ed7635dee49ba7f280[…]
UPDATE:
Found lines marked as INFO rather than error and this might be the reason I guess:
I, [2023-04-06T18:44:02.460059 #1] INFO -- : > cd /var/www/discourse && su discourse -c 'LOAD_PLUGINS=0 bundle exec rake plugin:pull_compatible_all'
I, [2023-04-06T18:44:04.581741 #1] INFO -- : docker_manager is already at latest compatible version
I, [2023-04-06T18:44:04.581929 #1] INFO -- : > cd /var/www/discourse && su discourse -c 'bundle exec rake db:migrate'
Discourse requires Redis 6.2.0 or up
I, [2023-04-06T18:44:05.937604 #1] INFO -- :
Yes, that’s exactly what I am doing, thank you. I only wish that message was marked as ERROR rather than INFO (among many other INFOs), and stopped right away. Lost a lot of time debugging all sort of things and not noticing this line