Helm Chart - monitor/ exporter issues

[stefanjay]

I use the default helm chart with following coniguration:

      envVariables:
        - name: GCPE_INTERNAL_MONITORING_LISTENER_ADDRESS
          value: tcp://127.0.0.1:8082

When I try the CLI in the pod:

/ $ gitlab-ci-pipelines-exporter monitor

FATA[0000]                                               error="rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing: dial tcp: address tcp://127.0.0.1:8082: too many colons in address\""

Also the metrics did not work but the access_token is correct and tested with curl against the gitlab api.

      config:
        # # Complete configuration syntax reference available here:
        # # https://github.com/mvisonneau/gitlab-ci-pipelines-exporter/blob/master/docs/configuration_syntax.md
        gitlab:
          url: http://gitlab-webservice-default.gitlab.svc.cluster.local:8080/
          token: glpat-7bqGyWPodxxxxxxxx
          enable_tls_verify: false

          health_url: http://gitlab-webservice-default.gitlab.svc.cluster.local:8080/-/health
          enable_health_check: true
        # projects:
        #   - name: gitlab-instance-89f24a72/gitlab-exporter-test-project
        #   - name: gitlab-instance-89f24a72/Monitoring
        wildcards:
          - owner:
              name: root
              kind: user

I added a scrape config to prometheus but get no metrics.

[stefanjay]

Update: It looks like the exporter throws an 403 when the gitlab project isn´t private. When the gitlab project is set anything else than private, the exporter fails. Is that intended?

[tboevil]

grpc only need host, no url. like this

[ouhzee]

grpc only need host, no url. like this

I'm still encountering this issue using binary. Is there any workaround for this @mvisonneau ?