gitlab-ce_16.9.0-ce.0_amd64.deb,HTTP 502

Hi friends, has anyone encountered this problem? My services are all normal, but a 502 error is displayed when opening the web. Does anyone know where to configure it?

run: alertmanager: (pid 1736412) 1368s; run: log: (pid 1736310) 1368s
run: gitaly: (pid 1695606) 1463s; run: log: (pid 3249552) 15543s
run: gitlab-exporter: (pid 1737797) 1366s; run: log: (pid 7060) 70867s
run: gitlab-kas: (pid 3294232) 13689s, got TERM; down: log: 0s, normally up, want up
run: gitlab-workhorse: (pid 1721807) 1402s; run: log: (pid 2116585) 20009s
run: logrotate: (pid 1735219) 1371s; run: log: (pid 7055) 70867s
run: nginx: (pid 1734903) 1371s; run: log: (pid 7054) 70867s
run: node-exporter: (pid 1735510) 1370s; run: log: (pid 7049) 70867s
run: postgres-exporter: (pid 1735983) 1369s; run: log: (pid 3347959) 11176s
run: postgresql: (pid 1737833) 1366s; run: log: (pid 7053) 70867s
run: prometheus: (pid 1737848) 1365s; run: log: (pid 3248130) 15583s
run: puma: (pid 1831201) 4s; run: log: (pid 3246742) 15649s
run: redis: (pid 1738031) 1364s; run: log: (pid 7062) 70867s
run: redis-exporter: (pid 1738054) 1363s; run: log: (pid 3250677) 15514s
run: sidekiq: (pid 1751816) 1332s; run: log: (pid 7050) 70867s

2024-02-19_16-48-19666×662 26.9 KB

Hey I have the same issue…were you able to solve the issue?

Which Linux Distribution, and host resources/hardware are used? Anything in the logs that could explain the server error 5xx and the delayed boot time (maybe a lock or database migration).

GitLab 16.9.1 was released today, might be worth a try to update and test, too.

im using alma linux 8.7, I basically am trying to deploy an instance of gitlab in terraform. It stands up a bunch of containers (gitlab-workhorse, shell, gitaly, pages, runner, zoekt indexer/webserver, redis, postgres, etc). And im getting that same error that tonywgx got, im looking at the logs of workhorse and this is what its saying.

{“correlation_id”:“”,“duration_ms”:51,“error”:“badgateway: failed to receive response: dial tcp: lookup webservice on 192.168.223.2:53: no such host”,“level”:“error”,“method”:“GET”,“msg”:“”,“time”:“2024-02-22T15:37:35Z”,“uri”:“”}

Thanks for the details. When you mention containers in this context, is this a custom Terraform module? Or, more refined question: Which documentation or tutorial are you following?

When SELinux is enabled, it might prevent file descriptor access causing problems. The syslog should be able to provide more insights.

well its a docker compose file that i found and im essentially converting it to terraform…

The docker-compose file in the Cloud Native container images project provides a development environment.. The images themselves are used in the Helm Chart and Kubernetes Operator to support cloud-native installation.

I would recommend following the official installation on Alma Linux, and use the Omnibus packages or container image instead of a custom setup. A custom setup is harder to support, introducing potential new problems and troubleshooting paths.

Thank you for the link.