GitLab

Removes the instance-specific 'monitor' label set on alerts, so that
multiple Prometheus instances in a HA setup all create the same
alerts.

Previously we were also counting 4xx errors, which are not relevant
as a reliability indicator. We may eventually want to restrict this
further to 502/503 (for backend timeouts only).

Caused by a typo.

Part of the optimization work on issue #51.

This metric takes quite a bit of evaluation time due to the large
cardinality of node_systemd_unit_state, and is unused.

Part of the optimization work in issue #51.

Display nice HTTP error pages where possible

See merge request !34

Add a favicon on all managed sites

See merge request !33

This just intercepts the /favicon.ico URL at the nginx level, if the
site defines its own (at a different URL), that one will be used instead.

Parameterize SSO ACL rules

See merge request !32

nginx: force cache directory ownership

See merge request !31

Turns out MemoryLimit= includes swap memory, not ideal.

This reverts commit 0ed7d159.

Nginx certbot

See merge request !30

ai3/prod#31

See also ai3/prod#31

mariadb: turn up log_warnings

See merge request !28

https://mariadb.com/kb/en/library/error-log/#configuring-the-error-log-verbosity

This reverts commit dd609e0e.

We are quite confident this won't fix the issue

Ignore port, the current version works only if the selected hostname
runs on the first port.