Fix nginx "single path" location blocks

Closes #293

See merge request archlinux/infrastructure!455

github-pull-closer[1] needs write access for closing the PRs.

[1] https://gitlab.archlinux.org/archlinux/github-pull-closer/

Fix #390

[1] https://github.blog/2021-04-05-behind-githubs-new-authentication-token-formats/

Ref #396

Clean up offload-build artifacts tasks and files

Closes #268

See merge request archlinux/infrastructure!478

A string is always "true"[1].

[1] https://stackoverflow.com/a/34868470

Cleanup tools

Closes #392

See merge request archlinux/infrastructure!485

Fix #392

install_arch: Use shell instead of command

See merge request archlinux/infrastructure!461

Thorben Günther authored 3 years ago and Kristian Klausen committed 3 years ago

shell is like command but runs the command through a shell, so that
piping will work.

Push docker logs to journald

Closes #326

See merge request !480

archwiki: ensure that MediaWiki file cache is invalidated on each MediaWiki update

See merge request !490

Jakub Klinkovský authored 3 years ago and Kristian Klausen committed 3 years ago

We've hit this problem at least twice in the past, last time with the new
Vector skin update. With this change it should not be necessary to do
anything manually after an update.

Install OOM configuration

See merge request !491

Jan Alexander Steffens (heftig) authored 3 years ago and Kristian Klausen committed 3 years ago

- When loki hits 5.5G RSS, aggressively reclaim its memory. This puts
  it under memory pressure.
- When loki hits 6G RSS, invoke the kernel OOM killer.
- When loki is under severe memory pressure, have systemd-oomd kill it.

Jan Alexander Steffens (heftig) authored 3 years ago and Kristian Klausen committed 3 years ago

- Allow killing of any cgroup if swap is about to run out. When swap is
  at least 90% full, oomd will kill a cgroup, tending to kill the cgroup
  with the highest swap usage.
- Allow killing cgroups of logged in users under severe memory pressure.
  When a cgroup is stalled waiting for memory at least 60% of the time,
  for at least 30 seconds, oomd will attempt to kill it or one of its
  descendant cgroups.

This partially reverts commit c3d00264.

The smartd changes are not ready to land yet and were included in the
above commit by accident.

phrik has a polkit rule for demize to restart phrik

Also adjust indentation and make the uploaded files owned by nobody.

Some people love to do 48 req/s to our tracker which causes ~100% cpu.
For now we allow 5req/sec with a burst of 10 requests.