1

On a CentOS 7 machine that I manage myself, I am/was running Gitlab 12.0.1.

Before a few updates back, everything was buttery smooth but an update broke my installation that whenever I go to my front page, I get the Deploy in progress message.

I did some debugging. gitlab-ctl status states that everything is up and running except for alertmanager.

I did a restart by gitlab-ctl restart which states that everything has restarted but afterwards checking the status again, alertmanager can be seen as down again.

I did a:

  • gitlab-ctl stop
  • gitlab-ctl reconfigure
  • gitlab-ctl start

in that order and everything went fine; however, alertmanager goes down again immediately.

And gitlab-ctl tail alertmanager states the following:

2019-06-27_09:04:19.75237 level=warn ts=2019-06-27T09:04:19.751036139Z caller=cluster.go:154 component=cluster err="couldn't deduce an advertise address: no private IP found, explicit advertise addr not provided"
2019-06-27_09:04:19.75630 level=error ts=2019-06-27T09:04:19.756225537Z caller=main.go:202 msg="unable to initialize gossip mesh" err="create memberlist: Failed to get final advertise address: No private IP address found, and explicit IP not provided"
2019-06-27_09:04:20.79887 level=info ts=2019-06-27T09:04:20.798268353Z caller=main.go:177 msg="Starting Alertmanager" version="(version=0.16.2, branch=master, revision=)"
2019-06-27_09:04:20.79890 level=info ts=2019-06-27T09:04:20.798339248Z caller=main.go:178 build_context="(go=go1.11.10, user=GitLab-Omnibus, date=)"
2019-06-27_09:04:20.80945 level=warn ts=2019-06-27T09:04:20.806462267Z caller=cluster.go:154 component=cluster err="couldn't deduce an advertise address: no private IP found, explicit advertise addr not provided"
2019-06-27_09:04:20.81497 level=error ts=2019-06-27T09:04:20.814894348Z caller=main.go:202 msg="unable to initialize gossip mesh" err="create memberlist: Failed to get final advertise address: No private IP address found, and explicit IP not provided"
2019-06-27_09:04:21.85821 level=info ts=2019-06-27T09:04:21.857774453Z caller=main.go:177 msg="Starting Alertmanager" version="(version=0.16.2, branch=master, revision=)"
2019-06-27_09:04:21.85823 level=info ts=2019-06-27T09:04:21.857842577Z caller=main.go:178 build_context="(go=go1.11.10, user=GitLab-Omnibus, date=)"
2019-06-27_09:04:21.86780 level=warn ts=2019-06-27T09:04:21.865566339Z caller=cluster.go:154 component=cluster err="couldn't deduce an advertise address: no private IP found, explicit advertise addr not provided"
2019-06-27_09:04:21.87528 level=error ts=2019-06-27T09:04:21.8751892Z caller=main.go:202 msg="unable to initialize gossip mesh" err="create memberlist: Failed to get final advertise address: No private IP address found, and explicit IP not provided"

So, obviously, there is something wrong but what it is.

p.s. Also, it should be noted that I'm not making use of the Nginx installation that comes packaged with gitlab-ce which is disabled anyways but I'm running my gitlab instance on my Apache installation. I restarted Apache somewhere along the way as well which yielded no errors.

2
  • 1
    I don't think that alertmanager actually is your problem... it's down for us as well but gitlab works fine without it. See gitlab.com/gitlab-org/omnibus-gitlab/issues/4556 however
    – Jörn Hees
    Commented Jul 29, 2019 at 17:31
  • @JörnHees Thanks for the pointer. Makes sense. Especially after the page you've provided. Guess alertmanager was a wild goose chase for me. My Gitlab installation works for me from the CLI to pull/push and just my web interface is down. Is it the same for you or your web interface is working? Commented Jul 30, 2019 at 8:20

1 Answer 1

3

Finally, I found out the solution. It’s really embarrassing but:

gitlab-ctl deploy-page status reported that the deploy page is up. Obviously.

Afterwards, I just ran gitlab-ctl deploy-page down and everything went back to normal. 😐

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .