platform-engineer-agent/k3s-cluster
1
0
Fork 0
forked from roger/k3s-cluster
Code Pull Requests Activity
139 Commits 18 Branches 0 Tags
a5291da0b248d53e90e658b08d13a20c5042b3da
Commit Graph

139 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
platform-engineer-agent
a5291da0b2 fix(gitea): bump restartedAt annotation to clear stuck gitea-runner pods (281 restarts from DNS resolution failures) 2026-07-21 15:36:57 +02:00
platform-engineer
075bdd8ca3 fix(gitea-runner): pin image tag to v0.7.2, add resource limits (256Mi/100m req, 512Mi/500m lim) to prevent OOM restarts 2026-07-20 22:19:37 +00:00
Roger Oriol
e44d7ba1fc Merge branch 'main' of https://git.rogi.casa/roger/k3s-cluster 2026-07-19 12:01:21 +02:00
Roger Oriol
f1005cd426 change openwebui url to ai.rogi.casa 2026-07-19 12:00:52 +02:00
Roger Oriol
43c0c2561e fix hermes write folder 2026-07-18 19:29:52 +02:00
roger
8334cd48f8 Merge pull request 'fix(myorg-assistant): add git-clone initContainer to git-sync CronJob' (#26) from fix/myorg-git-sync-clone into main 
Reviewed-on: roger/k3s-cluster#26
 2026-07-18 19:25:00 +02:00
Hermes Platform Engineer
ba7e05a73d fix(myorg-assistant): add git-clone initContainer to git-sync CronJob 
The git-sync CronJob only ran `git pull`/`git push` against /data/myorg
but never ensured the repo existed. The clone was solely the Deployment's
git-clone initContainer's job, and when that didn't populate the volume the
cron pod failed with 'Not a git repository: /data/myorg' on every run (and
swallowed the error, exiting 0).

Add an idempotent git-clone initContainer (guarded by [ ! -d /data/myorg/.git ])
so the cron job self-heals and actually syncs.

Refs: myorg-git-sync-* 'Pull: Error: Not a git repository' (last 24h+)
 2026-07-18 17:21:26 +00:00
Roger Oriol
8bc3025296 fix platform engineer not allowed to respond to discord messages 2026-07-18 19:20:23 +02:00
Roger Oriol
7a7d67bedc configure git token env variable in myorg assistant cronjobs 2026-07-18 19:03:46 +02:00
Roger Oriol
19cdc77880 add git token env var to deadline checker 2026-07-18 18:59:36 +02:00
Roger Oriol
8983f482d0 give more memory to homeassistant 2026-07-17 23:59:50 +02:00
Roger Oriol
0b27cefd13 myorg assistant cron jobs env variables 2026-07-15 00:32:23 +02:00
Roger Oriol
279cc1f235 configure litellm models 2026-07-14 21:45:18 +02:00
Roger Oriol
cf6e2784fe configure litellm models 2026-07-14 21:36:38 +02:00
Roger Oriol
dad38347e7 upgrade n8n memory requirements 2026-07-14 18:50:28 +02:00
Roger Oriol
a5b90994a4 fix myorg assistant 2026-07-11 19:22:57 +02:00
Roger Oriol
04b736287b fix argocd redirect 2026-07-09 23:48:47 +02:00
Roger Oriol
8c6950fd43 Merge branch 'main' of https://git.rogi.casa/roger/k3s-cluster 2026-07-09 00:45:40 +02:00
Roger Oriol
81dfe6fd60 fix argocd rbac and create agents.md 2026-07-09 00:45:10 +02:00
roger
5d80abf3e8 Merge pull request 'fix: myorg-assistant cronjobs ImagePullBackOff + argocd root drift loop' (#18) from platform-engineer-agent/k3s-cluster:fix-merge-to-main into main 
Reviewed-on: roger/k3s-cluster#18
 2026-07-09 00:22:54 +02:00
roger
48f18d2a3e Merge branch 'main' into fix-merge-to-main 2026-07-09 00:22:22 +02:00
Roger Oriol
ce08365e06 revert to qwen3.6 for platform engineer 2026-07-07 23:52:37 +02:00
Platform Engineer
0794153e56 fix(myorg-assistant): point cronjobs at registry image + add imagePullSecrets 
CronJob pods were stuck in ImagePullBackOff because they referenced
the local-only image 'myorg-assistant:latest' which is not present
on the node. Switch all 5 cronjobs to the Gitea registry image
git.rogi.casa/roger/myorg-assistant/myorg-assistant:fcf79bf
(matching the Deployment), set imagePullPolicy: Always, and add
imagePullSecrets: gitea-registry so they can authenticate to the
private registry.
 2026-07-06 14:11:44 +00:00
Platform Engineer
fc1b4383c1 fix(argocd): add ignoreDifferences to root app to stop Application CRD drift loop 
The k3s-cluster-root app-of-apps has been in a continuous Synced<->OutOfSync
oscillation (489 transitions in 24h). Each cycle, ArgoCD syncs the 'argocd'
Application CRD, succeeds, then immediately detects drift because ArgoCD
adds status/operation fields to Application resources at runtime.

Adding ignoreDifferences for /status and /operation on Application resources
stops the drift loop while keeping the app-of-apps functional.
 2026-07-06 14:11:44 +00:00
Roger Oriol
6b697c9665 configure higher cron schedules 2026-07-05 23:30:51 +02:00
Roger Oriol
08bb4de278 fix platform-engineer agent 2026-07-05 21:01:28 +02:00
Roger Oriol
2cccbc019f fix rbac role for platform engineer cron seed 2026-07-05 18:15:10 +02:00
Roger Oriol
3f29b77e55 remove k8s event exporter with bad sink 2026-07-05 17:48:38 +02:00
Roger Oriol
440fcf858f new argocd user platform-engineer 2026-07-05 17:39:54 +02:00
Roger Oriol
9fd7d02c7c loki + promtail new monitoring services 2026-07-05 17:32:15 +02:00
Roger Oriol
85c8cbfc31 fix searxng 2026-07-04 00:53:17 +02:00
Roger Oriol
9de2897f46 fix searxng 2026-07-04 00:51:11 +02:00
Roger Oriol
c3a07f75ab fix searxng 2026-07-04 00:44:49 +02:00
Roger Oriol
a567184347 fix searxng 2026-07-04 00:35:17 +02:00
Roger Oriol
54059cdb72 searxng 2026-07-04 00:19:48 +02:00
Roger Oriol
7faaa53855 searxng 2026-07-04 00:10:23 +02:00
Roger Oriol
6e689accd0 fix hermes cron 2026-06-29 00:48:23 +02:00
Roger Oriol
1145214e24 fix hermes qwen model 2026-06-29 00:21:57 +02:00
Roger Oriol
9eb8d344fa fix pihole liveness probe 2026-06-28 19:08:04 +02:00
Roger Oriol
22ef2a38b2 pihole unbound 2026-06-28 09:34:43 +02:00
Roger Oriol
d00c6fb63d fix qwen model name in hermes 2026-06-28 09:28:35 +02:00
Roger Oriol
734962d198 fix hermes liveness probe 2026-06-28 00:43:09 +02:00
Roger Oriol
4d9195b32d use hermes stock image for platform engineer 2026-06-27 21:00:03 +02:00
Roger Oriol
54579df4b3 use hermes stock image for platform engineer 2026-06-27 20:40:42 +02:00
Roger Oriol
3f3467cb13 gitea registry ingress 2026-06-27 11:46:53 +02:00
Roger Oriol
6e02d9a885 new platform engineer agent 2026-06-27 00:09:39 +02:00
Roger Oriol
d8012dfb6c monitoring: add dashboard ideas doc 
Survey of dashboards that could be built from existing and not-yet-enabled
metrics across the cluster's services (traefik, coredns, metallb, cert-manager,
phoenix, litellm, gitea, postgres, etc.), with per-service enable steps and
a recommended priority order.
 2026-06-26 20:22:54 +02:00
Roger Oriol
bf1387dc3e monitoring: add Grafana dashboards + kube-state-metrics & node-exporter 
Dashboards (provisioned via ConfigMaps into Grafana pod, 'K3s Cluster' folder):
- Cluster Overview: per-namespace CPU/mem/net/fs, pod counts, pod health (KSM)
- Pods & Services: per-pod CPU/mem/net/fs, throttling, pod status, restarts, PVCs
- Nodes: per-node CPU%/mem%, load average, disk usage, network (node-exporter)
- Control Plane & API Server: request rate, latency p95, 5xx, kubelet/PLEG
- Prometheus Self-Monitoring: ingestion, series, scrape duration, memory

Exporters (auto-scraped via existing kubernetes-service-endpoints job):
- kube-state-metrics: pod/deployment/PVC/replica state (kube_pod_status_phase,
  kube_pod_container_status_restarts_total, kube_persistentvolumeclaim_*)
- node-exporter (DaemonSet, hostNetwork): node_cpu_seconds_total,
  node_memory_*, node_filesystem_*, node_load*, node_network_*
 2026-06-26 19:48:17 +02:00
Roger Oriol
2eab82b430 fix nas ingress 2026-06-26 19:01:08 +02:00
Roger Oriol
3cdd40153f fix nas ingress 2026-06-26 18:54:17 +02:00