bug#60936: 30.0.50; ERC >5.5: Add erc-fill style based on visual-line-mode

Michael Albinus <[email protected]> writes:

> Michael Albinus <[email protected]> writes:
>
>> I've enabled the /metrics endpoint on emba. This requires a restart of
>> gitlab, which I haven't done. Should happen next time, when gitlab
>> patches are installed (which is not my responsibility).
>
> Hmm. I've just read <https://docs.gitlab.com/ee/operations/>. It tells us
>
> Measure reliability and stability with metrics (removed)
>
> This feature was deprecated in GitLab 14.7 and removed in 16.0.
>
> So it might work ATM (we're using GitLab CE 13.12.15), but it might
> disappear in the future.

Hi Michael. This deprecation notice appears to be about GitLab's metrics
feature, which they describe as a managed Prometheus instance and
integrated dashboard solution for their enterprise product. Apparently,
they're replacing that with a full observability offering. In case
you're curious, they also say [1]:

  This deprecation does not include:

  - Deprecating alerts for Prometheus
  - Capabilities that GitLab comes with that allow operators of GitLab
    to retrieve metrics from those instances [2]

It's the second bullet I was referring to initially, which allows
external Prometheus instances to poll the /-/metrics endpoint if
exposed. But to be of any use, those instances would need their IP
addresses whitelisted. Additionally, we'd need a "node exporter" [3]
process running on the same host to provide intel on system-resource
consumption. In the end, this is probably too involved to be worth
anyone's while. So, I guess you can probably just revert whatever change
you made to the configuration. Thanks anyway and please pardon the
distraction.

[1] https://gitlab.com/gitlab-org/gitlab/-/issues/346541
[2] https://docs.gitlab.com/ee/administration/monitoring/prometheus/gitlab_metrics.html
[3] https://prometheus.io/docs/guides/node-exporter/