Show newer

Yet Another Grafana Dashboard I hear you ask? YEAH

Monitoring time since last backup, and total backup size
stats.pixie.town/d/w63hEqeGk/b

also my workflow is quite the mess atm with like 2-3 editor windows and tons of terminals, extra browser windows and all workspace in use, but it kinda works? Both for coding and deploying my nixos stuff

Show thread

it's a shame borgmatic can't do hooks as well as just borg though :>
no way to get a per-repo finished hook with the name in some variable
(my old infra still runs with borgmatic, all the nixos stuff just uses borg, so you can see those in the graphs)

Show thread

Yet Another Grafana Dashboard I hear you ask? YEAH

Monitoring time since last backup, and total backup size
stats.pixie.town/d/w63hEqeGk/b

what the fuck why does it start to look like I know programming so much

i fucked up the git.pixie.town server and im rushing to try fix it but I have to go to dinner too >:(

next is writing a prometheus middleware that collects borg stats on backup completion, and alertmanager rules for when a backup didn't happen

Show thread

alertmanager for systemd jobs is amazing, seeing if a service crashes, or when a backup service fails

Show thread

definitely some rule tweaking to do still but it's going p well

Show thread

huh i have some decent alertmanager rules from awesome-prometheus-alerts.grep and alerting through Matrix with go-neb (with a NixOS module :DDD)

Show older
Pixietown

Small server part of the pixie.town infrastructure. Registration is closed.