Trying to upgrade from 12.3 to 12.10, I get only 502 status codes on any gitlab page.
Here are the errors found in the logs:
==> /var/log/gitlab/unicorn/unicorn_stderr.log <==
Settingslogic::MissingSetting: Missing setting 'action_cable' in /opt/gitlab/embedded/service/gitlab-rails/config/gitlab.yml
==> /var/log/gitlab/unicorn/unicorn_stdout.log <==
bundler: failed to load command: unicorn (/opt/gitlab/embedded/bin/unicorn)
==> /var/log/gitlab/gitlab-rails/production.log <==
Cached record for ApplicationSetting couldn't be loaded, falling back to uncached record: FATAL: the database system is shutting down
Cached record for ApplicationSetting couldn't be loaded, falling back to uncached record: FATAL: the database system is shutting down
Cached record for ApplicationSetting couldn't be loaded, falling back to uncached record: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket "/var/opt/gitlab/postgresql/.s.PGSQL.5432"?
Cached record for ApplicationSetting couldn't be loaded, falling back to uncached record: FATAL: the database system is starting up
==> /var/log/gitlab/postgresql/current <==
2021-11-11_11:26:03.25975 FATAL: the database system is shutting down
2021-11-11_11:26:03.49136 FATAL: the database system is shutting down
2021-11-11_11:26:03.49314 FATAL: the database system is shutting down
2021-11-11_11:26:12.34500 LOG: database system is shut down
2021-11-11_11:26:19.82447 LOG: listening on Unix socket "/var/opt/gitlab/postgresql/.s.PGSQL.5432"
2021-11-11_11:26:20.62921 FATAL: the database system is starting up
2021-11-11_11:26:20.63019 LOG: database system was shut down at 2021-11-11 11:26:09 GMT
2021-11-11_11:26:20.95011 FATAL: the database system is starting up
2021-11-11_11:26:20.96685 FATAL: the database system is starting up
2021-11-11_11:26:21.09774 LOG: database system is ready to accept connections
==> /var/log/gitlab/gitlab-workhorse/current <==
{"correlation_id":"JSGUSNWgPj1","duration_ms":0,"error":"badgateway: failed to receive response: dial unix /var/opt/gitlab/gitlab-rails/sockets/gitlab.socket: connect: no such file or directory","level":"error","method":"POST","msg":"error","time":"2021-11-11T11:27:53Z","uri":"/api/v4/jobs/request"}
==> /var/log/gitlab/gitlab-rails/sidekiq.log <==
Missing setting 'action_cable' in /opt/gitlab/embedded/service/gitlab-rails/config/gitlab.yml
==> /var/log/gitlab/sidekiq/current <==
Missing setting 'action_cable' in /opt/gitlab/embedded/service/gitlab-rails/config/gitlab.yml
However, everything seems to run fine:
# gitlab-ctl status
run: gitaly: (pid 3159) 2168s; run: log: (pid 2014) 41445417s
run: gitlab-workhorse: (pid 3169) 2167s; run: log: (pid 2029) 41445416s
run: logrotate: (pid 3182) 2167s; run: log: (pid 2019) 41445417s
run: nginx: (pid 3201) 2166s; run: log: (pid 2027) 41445416s
run: postgresql: (pid 3356) 2142s; run: log: (pid 2018) 41445417s
run: redis: (pid 3371) 2135s; run: log: (pid 2026) 41445416s
run: sidekiq: (pid 9035) 11s; run: log: (pid 2021) 41445416s
run: unicorn: (pid 9044) 10s; run: log: (pid 2022) 41445416s
When I check for further information, I get:
# gitlab-rake gitlab:check
rake aborted!
ArgumentError: could not find a temporary directory
I’m not able to set this action_cable setting properly.
I don’t know if there is actually a DB issue.
The socket is not created after a gitlab-ctl reconfigure
The /tmp directory has mode 777.
Any idea ?
Thanks