Compare commits

..
1 Commits
Author SHA1 Message Date
renovate-bot c985083f73 chore(deps): update renovate/renovate docker tag to v44.102.0
ci / lint-prettier (push) Successful in 11s
ci / lint-ruff (push) Successful in 5s
ci / lint-yaml (push) Successful in 7s
ci / deploy-userbot-panel (push) Has been skipped
ci / deploy-userbot-panel (pull_request) Has been skipped
ci / lint-dockerfiles (push) Successful in 4s
ci / validate (push) Successful in 5s
ci / lint-prettier (pull_request) Successful in 7s
ci / lint-ruff (pull_request) Successful in 4s
ci / lint-yaml (pull_request) Successful in 7s
ci / lint-dockerfiles (pull_request) Successful in 5s
ci / validate (pull_request) Successful in 6s
renovate-ci / validate-renovate (pull_request) Successful in 10s
ci / build (push) Has been skipped
ci / build (pull_request) Has been skipped
2026-09-18 16:18:01 +00:00
26 changed files with 52 additions and 703 deletions

No files matched your search

+1 -1
View File
@@ -48,5 +48,5 @@ jobs:
docker run --rm \
-v "$PWD:/work" \
-w /work \
renovate/renovate:44.97.2 \
renovate/renovate:44.83.2 \
renovate-config-validator renovate.json
-52
View File
@@ -1,52 +0,0 @@
name: renovate-run
on:
workflow_dispatch:
inputs:
repositories:
description: "Repositories to scan (comma-separated)"
required: false
default: "forust/homelab"
log_level:
description: "Renovate log level"
required: false
default: "info"
type: choice
options:
- info
- debug
concurrency:
group: renovate-run
cancel-in-progress: false
jobs:
run-renovate:
runs-on: [self-hosted, linux, arch, homelab]
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Run Renovate
shell: bash
env:
RENOVATE_TOKEN: ${{ secrets.RENOVATE_TOKEN }}
RENOVATE_GITHUB_COM_TOKEN: ${{ secrets.RENOVATE_GITHUB_COM_TOKEN }}
RENOVATE_REPOSITORIES: ${{ inputs.repositories }}
LOG_LEVEL: ${{ inputs.log_level }}
run: |
set -euo pipefail
: "${RENOVATE_TOKEN:?missing RENOVATE_TOKEN secret — add a renovate-bot PAT in repo/org Actions secrets}"
docker run --rm \
-v "$PWD/renovate/config.js:/opt/renovate/config.js:ro" \
-e RENOVATE_PLATFORM=gitea \
-e RENOVATE_ENDPOINT=https://gitea.forust.xyz/api/v1 \
-e RENOVATE_TOKEN="$RENOVATE_TOKEN" \
-e RENOVATE_GITHUB_COM_TOKEN="${RENOVATE_GITHUB_COM_TOKEN:-}" \
-e RENOVATE_REPOSITORIES="${RENOVATE_REPOSITORIES:-forust/homelab}" \
-e RENOVATE_CONFIG_FILE=/opt/renovate/config.js \
-e RENOVATE_BASE_DIR=/tmp/renovate \
-e LOG_LEVEL="${LOG_LEVEL:-info}" \
renovate/renovate:44.97.2
-15
View File
@@ -138,21 +138,6 @@ jobs:
--values "$repo/prometheus-stack/k8s/grafana-values.yaml" \
--wait
fi
if [ -f "$repo/loki/k8s/active" ]; then
echo "== Upgrading loki/alloy =="
helm repo add grafana https://grafana.github.io/helm-charts >/dev/null 2>&1 || true
helm repo update grafana >/dev/null 2>&1 || true
helm upgrade --install loki grafana/loki \
--version 7.3.0 \
--namespace prometheus \
--values "$repo/loki/k8s/loki-values.yaml" \
--wait
helm upgrade --install alloy grafana/alloy \
--version 1.12.1 \
--namespace prometheus \
--values "$repo/loki/k8s/alloy-values.yaml" \
--wait
fi
if [ "${#other_files[@]}" -gt 0 ]; then
echo " resources: ${other_files[*]}"
-1
View File
@@ -1 +0,0 @@
secret.yaml
View File
Whitespace-only changes.
-37
View File
@@ -1,37 +0,0 @@
apiVersion: apps/v1
kind: Deployment
metadata:
name: cloudflared
labels:
app: cloudflared
spec:
replicas: 1
selector:
matchLabels:
app: cloudflared
template:
metadata:
labels:
app: cloudflared
spec:
containers:
- name: cloudflared
image: cloudflare/cloudflared:2026.1.1
imagePullPolicy: IfNotPresent
args:
- tunnel
- --no-autoupdate
- run
env:
- name: TUNNEL_TOKEN
valueFrom:
secretKeyRef:
name: cloudflared-secrets
key: TUNNEL_TOKEN
resources:
requests:
memory: "32Mi"
cpu: "30m"
limits:
memory: "128Mi"
cpu: "200m"
-7
View File
@@ -1,7 +0,0 @@
apiVersion: v1
kind: Secret
metadata:
name: cloudflared-secrets
type: Opaque
stringData:
TUNNEL_TOKEN: your_tunnel_token_here
+35 -80
View File
@@ -1,102 +1,57 @@
container_runtime: containerd
agent:
acquisition: []
additionalAcquisition:
- labels:
type: traefik
limit: 1000
query: |
{namespace="traefik"}
source: loki
url: http://loki.prometheus.svc.cluster.local:3100/
wait_for_ready: 30s
env:
- name: COLLECTIONS
value: crowdsecurity/traefik crowdsecurity/base-http-scenarios
value: "crowdsecurity/traefik crowdsecurity/base-http-scenarios"
- name: DISABLE_COLLECTIONS
value: crowdsecurity/sshd
metrics:
enabled: true
serviceMonitor:
additionalLabels:
release: prometheus-stack
enabled: true
# Static machine identity: agent pods mount pre-created LAPI credentials
# (Secret crowdsec-agent-credentials, key local_api_credentials.yaml)
# at the exact path the agent entrypoint expects. Together with the
# patched register-init (enforced by janitor-cronjob.yaml) the agent
# never calls `cscli lapi register` in steady state, so pod names,
# restarts and reboots can no longer break it.
extraVolumes:
- name: static-creds
secret:
secretName: crowdsec-agent-credentials
items:
- key: local_api_credentials.yaml
path: local_api_credentials.yaml
extraVolumeMounts:
- name: static-creds
mountPath: /tmp_config/local_api_credentials.yaml
subPath: local_api_credentials.yaml
readOnly: true
value: "crowdsecurity/linux crowdsecurity/sshd"
acquisition:
- namespace: traefik
podName: "*traefik*"
program: traefik
poll_without_inotify: true
resources:
limits:
cpu: 200m
memory: 500Mi
requests:
cpu: 50m
memory: 100Mi
config:
parsers:
s02-enrich:
mobile-whitelist.yaml: |
name: forust/mobile-whitelist
description: "Whitelist SWAN/4ka mobile network"
whitelist:
reason: "Mobile IP whitelist"
cidr:
- "84.245.64.0/18"
postoverflows:
s01-whitelist:
home-dynamic-ip.yaml: |
name: forust/home-dynamic-ip
description: "Whitelist home dynamic IP"
whitelist:
reason: "Home dynamic IP"
expression:
- evt.Overflow.Alert.Source.IP in LookupHost("ddns.forust.xyz")
limits:
cpu: 200m
memory: 500Mi
lapi:
env:
- name: COLLECTIONS
value: crowdsecurity/traefik crowdsecurity/base-http-scenarios
value: "crowdsecurity/traefik crowdsecurity/base-http-scenarios"
- name: DISABLE_COLLECTIONS
value: crowdsecurity/linux crowdsecurity/sshd
metrics:
enabled: true
serviceMonitor:
additionalLabels:
release: prometheus-stack
enabled: true
value: "crowdsecurity/linux crowdsecurity/sshd"
service:
type: ClusterIP
persistentVolume:
config:
enabled: true
size: 100Mi
storageClassName: local-path-retain
data:
enabled: true
size: 1Gi
storageClassName: local-path-retain
size: 1Gi
config:
enabled: true
storageClassName: local-path-retain
size: 100Mi
storeLAPICscliCredentialsInSecret: true
resources:
limits:
cpu: 400m
memory: 500Mi
requests:
cpu: 50m
memory: 150Mi
service:
type: ClusterIP
storeLAPICscliCredentialsInSecret: true
limits:
cpu: 400m
memory: 500Mi
metrics:
enabled: true
serviceMonitor:
additionalLabels:
release: prometheus-stack
enabled: true
interval: 30s
scrapeTimeout: 10s
namespace: prometheus
-32
View File
@@ -1,32 +0,0 @@
apiVersion: v1
data:
crowdsec-overview.json: "{\n \"__inputs\": [\n {\n \"name\": \"DS_PROMETHEUS\",\n \"label\": \"Prometheus\",\n \"description\": \"\",\n \"type\": \"datasource\",\n \"pluginId\": \"prometheus\",\n \"pluginName\": \"Prometheus\"\n }\n ],\n \"__requires\": [\n {\n \"type\": \"grafana\",\n \"id\": \"grafana\",\n \"name\": \"Grafana\",\n \"version\": \"8.1.2\"\n },\n {\n \"type\": \"panel\",\n \"id\": \"graph\",\n \"name\": \"Graph (old)\",\n \"version\": \"\"\n },\n {\n \"type\": \"datasource\",\n \"id\": \"prometheus\",\n \"name\": \"Prometheus\",\n \"version\": \"1.0.0\"\n },\n {\n \"type\": \"panel\",\n \"id\": \"stat\",\n \"name\": \"Stat\",\n \"version\": \"\"\n },\n {\n \"type\": \"panel\",\n \"id\": \"timeseries\",\n \"name\": \"Time series\",\n \"version\": \"\"\n }\n ],\n \"annotations\": {\n \"list\": [\n {\n \"builtIn\": 1,\n \"datasource\": \"-- Grafana --\",\n \"enable\": true,\n \"hide\": true,\n \"iconColor\": \"rgba(0, 211, 255, 1)\",\n \"name\": \"Annotations & Alerts\",\n \"target\": {\n \"limit\": 100,\n \"matchAny\": false,\n \"tags\": [],\n \"type\": \"dashboard\"\n },\n \"type\": \"dashboard\"\n }\n ]\n },\n \"editable\": true,\n \"gnetId\": null,\n \"graphTooltip\": 0,\n \"id\": null,\n \"links\": [],\n \"panels\": [\n {\n \"collapsed\": false,\n \"datasource\": null,\n \"gridPos\": {\n \"h\": 1,\n \"w\": 24,\n \"x\": 0,\n \"y\": 0\n },\n \"id\": 24,\n \"panels\": [],\n \"title\": \"Summary\",\n \"type\": \"row\"\n },\n {\n \"cacheTimeout\": null,\n \"datasource\": \"${DS_PROMETHEUS}\",\n \"fieldConfig\": {\n \"defaults\": {\n \"color\": {\n \"mode\": \"thresholds\"\n },\n \"mappings\": [\n {\n \"options\": {\n \"match\": \"null\",\n \"result\": {\n \"text\": \"N/A\"\n }\n },\n \"type\": \"special\"\n }\n ],\n \"thresholds\": {\n \"mode\": \"absolute\",\n \"steps\": [\n {\n \"color\": \"#E02F44\",\n \"value\": null\n },\n {\n \"color\": \"#E02F44\",\n \"value\": 10\n },\n {\n \"color\": \"#299c46\",\n \"value\": 10\n }\n ]\n },\n \"unit\": \"none\"\n },\n \"overrides\": []\n },\n \"gridPos\": {\n \"h\": 8,\n \"w\": 6,\n \"x\": 0,\n \"y\": 1\n },\n \"id\": 2,\n \"interval\": null,\n \"links\": [],\n \"maxDataPoints\": 100,\n \"options\": {\n \"colorMode\": \"background\",\n \"graphMode\": \"none\",\n \"justifyMode\": \"auto\",\n \"orientation\": \"horizontal\",\n \"reduceOptions\": {\n \"calcs\": [\n \"lastNotNull\"\n ],\n \"fields\": \"\",\n \"values\": false\n },\n \"text\": {},\n \"textMode\": \"auto\"\n },\n \"pluginVersion\": \"8.1.2\",\n \"targets\": [\n {\n \"exemplar\": true,\n \"expr\": \"count(cs_info)\",\n \"interval\": \"\",\n \"legendFormat\": \"\",\n \"refId\": \"A\"\n }\n ],\n \"timeFrom\": null,\n \"timeShift\": null,\n \"title\": \"Running Crowdsec\",\n \"transparent\": true,\n \"type\": \"stat\"\n },\n {\n \"aliasColors\": {},\n \"bars\": false,\n \"dashLength\": 10,\n \"dashes\": false,\n \"datasource\": \"${DS_PROMETHEUS}\",\n \"decimals\": 1,\n \"fieldConfig\": {\n \"defaults\": {\n \"links\": []\n },\n \"overrides\": []\n },\n \"fill\": 1,\n \"fillGradient\": 0,\n \"gridPos\": {\n \"h\": 8,\n \"w\": 18,\n \"x\": 6,\n \"y\": 1\n },\n \"hiddenSeries\": false,\n \"id\": 8,\n \"legend\": {\n \"alignAsTable\": true,\n \"avg\": false,\n \"current\": false,\n \"max\": false,\n \"min\": false,\n \"rightSide\": true,\n \"show\": true,\n \"sort\": \"total\",\n \"sortDesc\": true,\n \"total\": true,\n \"values\": true\n },\n \"lines\": true,\n \"linewidth\": 1,\n \"nullPointMode\": \"null\",\n \"options\": {\n \"alertThreshold\": true\n },\n \"percentage\": false,\n \"pluginVersion\": \"8.1.2\",\n \"pointradius\": 2,\n \"points\": false,\n \"renLine truncated
kind: ConfigMap
metadata:
labels:
app.kubernetes.io/managed-by: manual
grafana_dashboard: "1"
name: crowdsec-crowdsec-overview
namespace: prometheus
---
apiVersion: v1
data:
crowdsec-lapi-metrics.json: "{\n \"__inputs\": [\n {\n \"name\": \"DS_PROMETHEUS\",\n \"label\": \"Prometheus\",\n \"description\": \"\",\n \"type\": \"datasource\",\n \"pluginId\": \"prometheus\",\n \"pluginName\": \"Prometheus\"\n }\n ],\n \"__requires\": [\n {\n \"type\": \"panel\",\n \"id\": \"bargauge\",\n \"name\": \"Bar gauge\",\n \"version\": \"\"\n },\n {\n \"type\": \"grafana\",\n \"id\": \"grafana\",\n \"name\": \"Grafana\",\n \"version\": \"8.1.2\"\n },\n {\n \"type\": \"datasource\",\n \"id\": \"prometheus\",\n \"name\": \"Prometheus\",\n \"version\": \"1.0.0\"\n }\n ],\n \"annotations\": {\n \"list\": [\n {\n \"builtIn\": 1,\n \"datasource\": \"-- Grafana --\",\n \"enable\": true,\n \"hide\": true,\n \"iconColor\": \"rgba(0, 211, 255, 1)\",\n \"name\": \"Annotations & Alerts\",\n \"target\": {\n \"limit\": 100,\n \"matchAny\": false,\n \"tags\": [],\n \"type\": \"dashboard\"\n },\n \"type\": \"dashboard\"\n }\n ]\n },\n \"editable\": true,\n \"gnetId\": null,\n \"graphTooltip\": 0,\n \"id\": null,\n \"iteration\": 1655915193937,\n \"links\": [],\n \"panels\": [\n {\n \"collapsed\": false,\n \"datasource\": null,\n \"gridPos\": {\n \"h\": 1,\n \"w\": 24,\n \"x\": 0,\n \"y\": 0\n },\n \"id\": 10,\n \"panels\": [],\n \"title\": \"Agents\",\n \"type\": \"row\"\n },\n {\n \"datasource\": \"${DS_PROMETHEUS}\",\n \"fieldConfig\": {\n \"defaults\": {\n \"color\": {\n \"mode\": \"thresholds\"\n },\n \"mappings\": [],\n \"thresholds\": {\n \"mode\": \"absolute\",\n \"steps\": [\n {\n \"color\": \"green\",\n \"value\": null\n },\n {\n \"color\": \"red\",\n \"value\": 80\n }\n ]\n }\n },\n \"overrides\": []\n },\n \"gridPos\": {\n \"h\": 8,\n \"w\": 12,\n \"x\": 0,\n \"y\": 1\n },\n \"id\": 2,\n \"options\": {\n \"displayMode\": \"gradient\",\n \"orientation\": \"vertical\",\n \"reduceOptions\": {\n \"calcs\": [\n \"lastNotNull\"\n ],\n \"fields\": \"\",\n \"values\": false\n },\n \"showUnfilled\": false,\n \"text\": {}\n },\n \"pluginVersion\": \"8.1.2\",\n \"repeat\": \"query0\",\n \"repeatDirection\": \"h\",\n \"targets\": [\n {\n \"exemplar\": false,\n \"expr\": \"sum(rate(cs_lapi_request_duration_seconds_bucket{endpoint=\\\"/v1/watchers/login\\\", instance=\\\"$lapi\\\"}[$__rate_interval])) by (le)\",\n \"format\": \"heatmap\",\n \"interval\": \"\",\n \"legendFormat\": \"{{le}}\",\n \"refId\": \"A\"\n }\n ],\n \"title\": \"Agents Login\",\n \"type\": \"heatmap\"\n },\n {\n \"datasource\": \"${DS_PROMETHEUS}\",\n \"fieldConfig\": {\n \"defaults\": {\n \"color\": {\n \"mode\": \"thresholds\"\n },\n \"mappings\": [],\n \"thresholds\": {\n \"mode\": \"absolute\",\n \"steps\": [\n {\n \"color\": \"green\",\n \"value\": null\n }\n ]\n },\n \"unit\": \"none\"\n },\n \"overrides\": []\n },\n \"gridPos\": {\n \"h\": 8,\n \"w\": 12,\n \"x\": 12,\n \"y\": 1\n },\n \"id\": 6,\n \"options\": {\n \"displayMode\": \"gradient\",\n \"orientation\": \"auto\",\n \"reduceOptions\": {\n \"calcs\": [\n \"lastNotNull\"\n ],\n \"fields\": \"\",\n \"values\": false\n },\n \"showUnfilled\": false,\n \"text\": {}\n },\n \"pluginVersion\": \"8.1.2\",\n \"targets\": [\n {\n \"exemplar\": true,\n \"expr\": \"sum(rate(cs_lapi_request_duration_seconds_bucket{endpoint=\\\"/v1/watchers/login\\\"}[$__rate_interval])) by (le)\",\n \"format\": \"heatmap\",\n \"interval\": \"\",\n \"legendFormat\": \"{{le}}\",\n \"refId\": \"A\"\n }\n ],\n \"title\": \"Heartbeat\",\n \"type\": \"heatmap\"\n },\n {\n \"collapsed\": false,\n \"datasource\": null,\n \"gridPos\": {\n \"h\": 1,\n \"w\": 24,\n \"x\": 0,\n \"y\": 9\n },\n \"id\": 12,\n \"panels\": [],\n \"title\": \"Decisions\",\n \"type\": \"row\"\n },\n {\n \"datasource\": \"${DS_PROMETHEUS}\",\n Line truncated
kind: ConfigMap
metadata:
labels:
app.kubernetes.io/managed-by: manual
grafana_dashboard: "1"
name: crowdsec-crowdsec-lapi-metrics
namespace: prometheus
---
apiVersion: v1
data:
crowdsec-insight.json: "{\n \"__inputs\": [\n {\n \"name\": \"DS_PROMETHEUS\",\n \"label\": \"Prometheus\",\n \"description\": \"\",\n \"type\": \"datasource\",\n \"pluginId\": \"prometheus\",\n \"pluginName\": \"Prometheus\"\n }\n ],\n \"__requires\": [\n {\n \"type\": \"panel\",\n \"id\": \"bargauge\",\n \"name\": \"Bar gauge\",\n \"version\": \"\"\n },\n {\n \"type\": \"panel\",\n \"id\": \"gauge\",\n \"name\": \"Gauge\",\n \"version\": \"\"\n },\n {\n \"type\": \"grafana\",\n \"id\": \"grafana\",\n \"name\": \"Grafana\",\n \"version\": \"8.1.2\"\n },\n {\n \"type\": \"datasource\",\n \"id\": \"prometheus\",\n \"name\": \"Prometheus\",\n \"version\": \"1.0.0\"\n },\n {\n \"type\": \"panel\",\n \"id\": \"stat\",\n \"name\": \"Stat\",\n \"version\": \"\"\n }\n ],\n \"annotations\": {\n \"list\": [\n {\n \"builtIn\": 1,\n \"datasource\": \"-- Grafana --\",\n \"enable\": true,\n \"hide\": true,\n \"iconColor\": \"rgba(0, 211, 255, 1)\",\n \"name\": \"Annotations & Alerts\",\n \"target\": {\n \"limit\": 100,\n \"matchAny\": false,\n \"tags\": [],\n \"type\": \"dashboard\"\n },\n \"type\": \"dashboard\"\n }\n ]\n },\n \"editable\": true,\n \"gnetId\": null,\n \"graphTooltip\": 0,\n \"id\": null,\n \"iteration\": 1655915159751,\n \"links\": [],\n \"panels\": [\n {\n \"collapsed\": true,\n \"datasource\": null,\n \"gridPos\": {\n \"h\": 1,\n \"w\": 24,\n \"x\": 0,\n \"y\": 0\n },\n \"id\": 22,\n \"panels\": [\n {\n \"cacheTimeout\": null,\n \"datasource\": \"${DS_PROMETHEUS}\",\n \"fieldConfig\": {\n \"defaults\": {\n \"color\": {\n \"mode\": \"thresholds\"\n },\n \"mappings\": [\n {\n \"options\": {\n \"match\": \"null\",\n \"result\": {\n \"text\": \"N/A\"\n }\n },\n \"type\": \"special\"\n }\n ],\n \"thresholds\": {\n \"mode\": \"absolute\",\n \"steps\": [\n {\n \"color\": \"green\",\n \"value\": null\n },\n {\n \"color\": \"red\",\n \"value\": 80\n }\n ]\n },\n \"unit\": \"dateTimeAsIso\"\n },\n \"overrides\": []\n },\n \"gridPos\": {\n \"h\": 9,\n \"w\": 5,\n \"x\": 2,\n \"y\": 1\n },\n \"id\": 2,\n \"interval\": null,\n \"links\": [],\n \"maxDataPoints\": 100,\n \"options\": {\n \"colorMode\": \"none\",\n \"graphMode\": \"none\",\n \"justifyMode\": \"auto\",\n \"orientation\": \"horizontal\",\n \"reduceOptions\": {\n \"calcs\": [\n \"lastNotNull\"\n ],\n \"fields\": \"\",\n \"values\": false\n },\n \"text\": {},\n \"textMode\": \"auto\"\n },\n \"pluginVersion\": \"8.1.2\",\n \"targets\": [\n {\n \"exemplar\": true,\n \"expr\": \"(process_start_time_seconds{instance=\\\"$instance\\\"})*1000\",\n \"interval\": \"\",\n \"legendFormat\": \"{{instance}}\",\n \"refId\": \"A\"\n }\n ],\n \"timeFrom\": null,\n \"timeShift\": null,\n \"title\": \"Up since\",\n \"type\": \"stat\"\n },\n {\n \"datasource\": \"${DS_PROMETHEUS}\",\n \"fieldConfig\": {\n \"defaults\": {\n \"displayName\": \"\",\n \"mappings\": [],\n \"thresholds\": {\n \"mode\": \"absolute\",\n \"steps\": [\n {\n \"color\": \"green\",\n \"value\": null\n }\n ]\n },\n \"unit\": \"decbytes\"\n },\n \"overrides\": []\n },\n \"gridPos\": {\n \"h\": 9,\n \"w\": 5,\n \"x\": 7,\n \"y\": 1\n },\n \"id\": 4,\n \"options\": {\n \"orientation\": \"auto\",\n \"reduceOptions\": {\n \"calcs\": [\n \"mean\"\n ],\n \"fields\": \"\",\n \"values\": false\n },\Line truncated
kind: ConfigMap
metadata:
labels:
app.kubernetes.io/managed-by: manual
grafana_dashboard: "1"
name: crowdsec-crowdsec-insight
namespace: prometheus
-195
View File
@@ -1,195 +0,0 @@
# CrowdSec self-healing: static machine identity + enforcement loops.
#
# Problem it fixes: the chart's agent init container runs
# `cscli lapi register --machine "$POD_NAME" ...`
# unconditionally. Credentials live in an emptyDir, the machine row lives
# in LAPI's persistent DB. Any init re-run for an already-known pod name
# (kubelet restart, node reboot) dies with
# 403 Forbidden: user '<pod>' already exist
# and the DaemonSet pod sticks in Init forever. Every DS restart also
# leaves an orphan machine row that is never cleaned.
#
# Design (name-independent):
# * Agent identity is a STATIC machine `crowdsec-agent-workstation`
# whose password lives in Secret `crowdsec-agent-credentials`
# (created once, manually - like all other secrets in this repo).
# The secret is mounted into agent pods at
# /tmp_config/local_api_credentials.yaml (see extraVolumeMounts in
# crowdsec-values.yaml), which is exactly the path the agent's main
# container copies into place at startup.
# * The DS init command is patched (strategic merge, by container name)
# to SKIP registration when that file exists, keeping the legacy
# register path only as fallback. Detection marker in the patched
# command: `[ -s /tmp_config`.
# * This CronJob enforces the desired state hourly, so recovery is
# automatic even after `helm upgrade` reverts the DS patch or the
# LAPI database is wiped:
# 1. patch DS init if it still has the unconditional register
# (no-op otherwise - no restart churn);
# 2. prune machines with no heartbeat for 2h (orphan hygiene);
# 3. ensure the static machine exists, recreating it with the
# Secret password if missing (agent retry loops reconnect
# on their own - same name + same password);
# 4. prune bouncer entries idle for 30d.
#
# Manual apply (crowdsec/k8s is NOT managed by deploy.yaml):
# kubectl apply -f crowdsec/k8s/janitor-cronjob.yaml
# Force a run:
# kubectl create job -n crowdsec --from=cronjob/crowdsec-janitor janitor-now
#
# Helm upgrades: the janitor's strategic patch puts the DS field under
# the `kubectl-patch` field manager, so a plain `helm upgrade` FAILS
# with an SSA conflict on initContainers[].command. Procedure:
# 1. revert init to chart state (kills the conflict):
# helm template crowdsec crowdsec/crowdsec --version <ver> \
# -n crowdsec -f crowdsec/k8s/crowdsec-values.yaml > /tmp/r.yaml
# python3 -c "import yaml,json; ..." # build revert patch from
# the rendered DaemonSet init command, then
# kubectl patch ds crowdsec-agent -n crowdsec \
# --type strategic -p "\$(cat /tmp/revert_patch.json)"
# 2. helm upgrade --install crowdsec ... (no --force needed)
# 3. janitor-now right away (upgrade reverts init; new pods would
# sit in Init until the next hourly run otherwise).
#
# One-time bootstrap (order matters):
# 1. Create Secret + static machine (see commands in chat).
# 2. Apply this file, trigger janitor-now, wait for agent 1/1.
# 3. One-time orphan cleanup:
# kubectl exec -n crowdsec deploy/crowdsec-lapi -- \
# cscli machines prune --duration 1h --force
# 4. Only then `helm upgrade` crowdsec with the extraVolumes values.
# Upgrade reverts the DS patch; trigger janitor-now right after it
# (otherwise new pods sit in Init until the next hourly run, then
# self-heal anyway).
#
# Password rotation: update the Secret, delete the machine
# (`cscli machines delete crowdsec-agent-workstation`), trigger
# janitor-now (recreates it), then `kubectl rollout restart
# ds/crowdsec-agent -n crowdsec` (agent reads the file at startup only).
apiVersion: v1
kind: ServiceAccount
metadata:
name: crowdsec-janitor
namespace: crowdsec
labels:
app.kubernetes.io/part-of: crowdsec
---
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
name: crowdsec-janitor
namespace: crowdsec
labels:
app.kubernetes.io/part-of: crowdsec
rules:
- apiGroups: [""]
resources: ["pods"]
verbs: ["get", "list"]
- apiGroups: [""]
resources: ["pods/exec"]
verbs: ["create"]
- apiGroups: ["apps"]
resources: ["daemonsets"]
verbs: ["get", "patch"]
# `kubectl exec deploy/<name>` resolves deploy -> replicaset -> pod,
# which needs read access to these (exec itself is pods/exec above).
- apiGroups: ["apps"]
resources: ["deployments", "replicasets"]
verbs: ["get", "list"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
name: crowdsec-janitor
namespace: crowdsec
labels:
app.kubernetes.io/part-of: crowdsec
subjects:
- kind: ServiceAccount
name: crowdsec-janitor
namespace: crowdsec
roleRef:
kind: Role
name: crowdsec-janitor
apiGroup: rbac.authorization.k8s.io
---
apiVersion: batch/v1
kind: CronJob
metadata:
name: crowdsec-janitor
namespace: crowdsec
labels:
app.kubernetes.io/part-of: crowdsec
spec:
schedule: "17 * * * *"
concurrencyPolicy: Forbid
successfulJobsHistoryLimit: 3
failedJobsHistoryLimit: 3
jobTemplate:
spec:
activeDeadlineSeconds: 300
template:
metadata:
labels:
app.kubernetes.io/part-of: crowdsec
spec:
serviceAccountName: crowdsec-janitor
restartPolicy: OnFailure
containers:
- name: janitor
# Same image the chart itself uses for registration jobs;
# IfNotPresent so it works while the node is offline
# (layer cached from the chart install).
image: alpine/kubectl:latest
imagePullPolicy: IfNotPresent
env:
- name: AGENT_PASSWORD
valueFrom:
secretKeyRef:
name: crowdsec-agent-credentials
key: password
command:
- /bin/sh
- -c
- |
set -eu
LAPI_EXEC="kubectl exec -n crowdsec deploy/crowdsec-lapi --"
echo "== 1. enforce patched agent init =="
CUR=$(kubectl get ds crowdsec-agent -n crowdsec \
-o jsonpath='{.spec.template.spec.initContainers[0].command[2]}')
case "$CUR" in
*'-s /tmp_config'*)
echo "init already patched"
;;
*)
echo "patching init"
WAIT='until nc "$LAPI_HOST" "$LAPI_PORT" -z'
WAIT="$WAIT; do echo waiting for lapi to start; sleep 5; done"
LINK='ln -s /staging/etc/crowdsec /etc/crowdsec'
REG='cscli lapi register --machine "$USERNAME"'
REG="$REG -u \"\$LAPI_URL\" --token \"\$REGISTRATION_TOKEN\""
CREDS=/tmp_config/local_api_credentials.yaml
CMD="$WAIT; $LINK; [ -s $CREDS ] || {"
CMD="$CMD $REG && cp"
CMD="$CMD /etc/crowdsec/local_api_credentials.yaml $CREDS; }"
ESC=$(printf '%s' "$CMD" | sed 's/"/\\"/g')
PATCH='{"spec":{"template":{"spec":{"initContainers":'
PATCH=$PATCH'[{"name":"wait-for-lapi-and-register",'
PATCH=$PATCH'"command":["sh","-c","'$ESC'"]}]}}}}'
kubectl patch ds crowdsec-agent -n crowdsec \
--type strategic -p "$PATCH"
;;
esac
echo "== 2. prune orphan machines (no heartbeat for 2h) =="
$LAPI_EXEC cscli machines prune --duration 2h --force
echo "== 3. ensure static machine exists =="
if $LAPI_EXEC cscli machines inspect \
crowdsec-agent-workstation >/dev/null 2>&1; then
echo "static machine present"
else
echo "recreating static machine"
$LAPI_EXEC cscli machines add crowdsec-agent-workstation \
--password "$AGENT_PASSWORD" --force
fi
echo "== 4. prune stale bouncers (no pull for 30d) =="
$LAPI_EXEC cscli bouncers prune -d 720h --force
-7
View File
@@ -25,10 +25,3 @@ spec:
ports:
- protocol: TCP
port: 8080
- from:
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: prometheus
ports:
- protocol: TCP
port: 6060
View File
Whitespace-only changes.
-57
View File
@@ -1,57 +0,0 @@
# Pinned chart: grafana/alloy 1.12.1 (app v1.19.2).
# Install (deferred to deploy task, namespace prometheus):
# helm upgrade --install alloy grafana/alloy --version 1.12.1 \
# --namespace prometheus --values loki/k8s/alloy-values.yaml --wait
# DaemonSet ships k8s pod logs (API-tailed, no hostPath mounts) to Loki.
# Scope phase 1: k8s only, compose leftovers out.
controller:
type: daemonset
resources:
requests:
memory: "128Mi"
cpu: "50m"
limits:
memory: "512Mi"
cpu: "500m"
image:
tag: "v1.19.2"
alloy:
configMap:
create: true
content: |
discovery.kubernetes "pods" {
role = "pod"
}
discovery.relabel "pods" {
targets = discovery.kubernetes.pods.targets
rule {
source_labels = ["__meta_kubernetes_namespace"]
target_label = "namespace"
}
rule {
source_labels = ["__meta_kubernetes_pod_name"]
target_label = "pod"
}
rule {
source_labels = ["__meta_kubernetes_pod_container_name"]
target_label = "container"
}
}
loki.source.kubernetes "pods" {
targets = discovery.relabel.pods.output
forward_to = [loki.write.default.receiver]
}
loki.write "default" {
endpoint {
url = "http://loki-gateway.prometheus.svc.cluster.local/loki/api/v1/push"
}
}
-97
View File
@@ -1,97 +0,0 @@
# Pinned chart: grafana/loki 7.3.0 (app 3.6.12).
# Install (deferred to deploy task, namespace prometheus):
# helm upgrade --install loki grafana/loki --version 7.3.0 \
# --namespace prometheus --values loki/k8s/loki-values.yaml --wait
# SingleBinary, filesystem storage on local-path-retain, 14d retention.
# No IngressRoute: Loki is cluster-internal, queried via Grafana datasource.
deploymentMode: SingleBinary
loki:
# Multitenancy off: single-node homelab, gateway + Alloy + Grafana talk to one tenant.
auth_enabled: false
image:
tag: "3.6.12"
commonConfig:
# Single replica: default RF=3 would require 3 ingesters and fail all writes.
replication_factor: 1
storage:
type: filesystem
schemaConfig:
configs:
- from: "2024-04-01"
store: tsdb
object_store: filesystem
schema: v13
index:
prefix: index_
period: 24h
compactor:
retention_enabled: true
delete_request_store: filesystem
limits_config:
retention_period: 336h
rulerConfig:
wal:
dir: /var/loki/ruler-wal
storage:
type: local
local:
directory: /var/loki/rules
singleBinary:
replicas: 1
persistence:
enabled: true
size: 20Gi
storageClass: local-path-retain
resources:
requests:
memory: "512Mi"
cpu: "200m"
limits:
memory: "2Gi"
cpu: "1000m"
# Zeroed: unused in SingleBinary mode (chart validation requires it).
write:
replicas: 0
read:
replicas: 0
backend:
replicas: 0
gateway:
replicas: 1
resources:
requests:
memory: "64Mi"
cpu: "50m"
limits:
memory: "256Mi"
cpu: "300m"
monitoring:
serviceMonitor:
enabled: true
labels:
release: prometheus-stack
interval: 15s
rules:
enabled: true
namespace: prometheus
labels:
release: prometheus-stack
# Disabled: memcached caches don't fit a memory-tight single node.
# SingleBinary works without them (slower repeated queries, fine at homelab scale).
resultsCache:
enabled: false
chunksCache:
enabled: false
# Disabled: synthetic canary traffic + helm test pod, noise on a single node.
lokiCanary:
enabled: false
test:
enabled: false
-5
View File
@@ -1,10 +1,5 @@
route:
receiver: default
routes:
- receiver: "null"
matchers:
- alertname="Watchdog"
receivers:
- name: default
- name: "null"
@@ -1,6 +1,3 @@
templates:
- "/etc/alertmanager/config/telegram.tmpl"
route:
receiver: telegram
group_by:
@@ -10,10 +7,10 @@ route:
group_interval: 5m
repeat_interval: 12h
routes:
- receiver: "null"
- receiver: null
matchers:
- alertname="InfoInhibitor"
- receiver: "null"
- receiver: null
matchers:
- alertname="Watchdog"
@@ -38,5 +35,4 @@ receivers:
chat_id: REPLACE_WITH_TELEGRAM_CHAT_ID
parse_mode: HTML
send_resolved: true
message: '{{ template "telegram.forust.message" . }}'
- name: "null"
- name: null
@@ -1,44 +0,0 @@
{{- define "telegram.forust.message" -}}
{{- $statusEmoji := "🚨" -}}
{{- if ne .Status "firing" -}}{{- $statusEmoji = "✅" -}}{{- end -}}
{{- $count := len .Alerts.Firing -}}
{{- if eq .Status "resolved" -}}{{- $count = len .Alerts.Resolved -}}{{- end -}}
{{ $statusEmoji }} <b>{{ .Status | toUpper }} ({{ $count }})</b>
{{- range .Alerts }}
<b>{{ .Labels.alertname }}</b>
{{- $sev := .Labels.severity }}
{{- if eq $sev "critical" }} 🔥 critical
{{- else if eq $sev "warning" }} ⚠️ warning
{{- else if eq $sev "info" }} ℹ️ info
{{- else if $sev }} • {{ $sev }}
{{- end }}
{{- if .Annotations.description }}
<i>{{ .Annotations.description }}</i>
{{- end }}
{{- if .Labels.namespace }}
📦 Namespace: <code>{{ .Labels.namespace }}</code>
{{- end }}
{{- if .Labels.pod }}
📦 Pod: <code>{{ .Labels.pod }}</code>
{{- end }}
{{- if .Labels.container }}
🐳 Container: <code>{{ .Labels.container }}</code>
{{- end }}
{{- if .Labels.node }}
🖥 Node: <code>{{ .Labels.node }}</code>
{{- end }}
{{- if .Labels.instance }}
🖥 Instance: <code>{{ .Labels.instance }}</code>
{{- end }}
{{- if .Labels.job }}
🔧 Job: <code>{{ .Labels.job }}</code>
{{- end }}
{{- if eq .Status "firing" }}
🕐 Since: <code>{{ .StartsAt | date "2006-01-02 15:04 MST" }}</code>
{{- else }}
🕐 Resolved: <code>{{ .EndsAt | date "2006-01-02 15:04 MST" }}</code>
{{- end }}
{{- end }}
{{- end }}
-18
View File
@@ -66,21 +66,3 @@ spec:
annotations:
summary: "Node memory pressure"
description: "Node {{ $labels.instance }} has used more than 90% of memory for 15 minutes."
- alert: LokiDown
expr: kube_statefulset_status_replicas_unavailable{statefulset="loki"} > 0 or kube_deployment_status_replicas_unavailable{deployment="loki-gateway"} > 0
for: 10m
labels:
severity: warning
annotations:
summary: "Loki is down"
description: "Loki in namespace {{ $labels.namespace }} has unavailable replicas for more than 10 minutes. Logs are not queryable."
- alert: AlloyDown
expr: kube_daemonset_status_number_unavailable{daemonset="alloy"} > 0
for: 10m
labels:
severity: warning
annotations:
summary: "Alloy is down"
description: "Alloy DaemonSet in namespace {{ $labels.namespace }} has {{ $value }} unavailable pods for more than 10 minutes. Pod logs are not being shipped to Loki."
+5 -16
View File
@@ -14,11 +14,8 @@ grafana:
persistence:
enabled: true
# Matches live PVC (10Gi/local-path). Retain migration is a separate task
# with data migration (see storage-audit doc); do NOT change SC/size here
# without migrating, helm upgrade fails on immutable PVC fields.
storageClassName: local-path
size: 10Gi
storageClassName: local-path-retain
size: 20Gi
ingress:
enabled: false
@@ -26,12 +23,6 @@ grafana:
service:
port: 80
additionalDataSources:
- name: Loki
type: loki
url: http://loki-gateway.prometheus.svc.cluster.local
access: proxy
prometheus:
prometheusSpec:
retention: 60d
@@ -39,8 +30,7 @@ prometheus:
storageSpec:
volumeClaimTemplate:
spec:
# Matches live PVC, see note on grafana.persistence above.
storageClassName: "local-path"
storageClassName: "local-path-retain"
accessModes:
- ReadWriteOnce
resources:
@@ -58,13 +48,12 @@ alertmanager:
storage:
volumeClaimTemplate:
spec:
# Matches live PVC (20Gi), see note on grafana.persistence above.
storageClassName: local-path
storageClassName: local-path-retain
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 20Gi
storage: 10Gi
defaultRules:
disabled:
+1
View File
@@ -12,6 +12,7 @@ spec:
middlewares:
- name: crowdsec-bouncer
namespace: crowdsec
- name: "security-chain@file"
services:
- name: prometheus-stack-grafana
port: 80
+1 -4
View File
@@ -1,10 +1,7 @@
{
"$schema": "https://docs.renovatebot.com/renovate-schema.json",
"extends": ["config:recommended"],
"enabledManagers": ["docker-compose", "kubernetes", "helm-values"],
"helm-values": {
"managerFilePatterns": ["/k8s/.+values\\.ya?ml$/"]
},
"enabledManagers": ["docker-compose", "kubernetes"],
"kubernetes": {
"managerFilePatterns": ["/k8s/.+\\.ya?ml$/"]
},
+1 -14
View File
@@ -24,25 +24,12 @@ The `renovate/k8s/active` marker makes the normal deployment workflow include
the namespace, ConfigMap, and CronJob. The Secret is intentionally excluded
from Git and must be applied separately after every new cluster.
Run it immediately instead of waiting for the six-hour schedule.
Two options, both use the same `renovate/config.js`:
Run it immediately instead of waiting for the six-hour schedule:
```sh
kubectl create job --from=cronjob/renovate renovate-manual-$(date +%s) -n renovate
```
or the `renovate-run` Actions workflow (Actions tab → `renovate-run` →
Run workflow). It runs `renovate/renovate:44.97.2` on the self-hosted
runner via Docker. Required Actions secrets (repo or org settings):
- `RENOVATE_TOKEN` — renovate-bot PAT (repository + issue read/write).
- `RENOVATE_GITHUB_COM_TOKEN` — optional, for changelogs and GitHub rate limits.
Inputs: `repositories` (default `forust/homelab`), `log_level`
(`info`/`debug`). Only one run at a time (concurrency group
`renovate-run`), same as the CronJob `Forbid` policy.
Inspect runs with:
```sh
+1 -4
View File
@@ -1,10 +1,7 @@
module.exports = {
platform: 'gitea',
endpoint: process.env.RENOVATE_ENDPOINT,
enabledManagers: ['docker-compose', 'kubernetes', 'helm-values'],
'helm-values': {
managerFilePatterns: ['/k8s/.+values\\.ya?ml$/'],
},
enabledManagers: ['docker-compose', 'kubernetes'],
kubernetes: {
managerFilePatterns: ['/k8s/.+\\.ya?ml$/'],
},
+2 -8
View File
@@ -8,10 +8,7 @@ data:
module.exports = {
platform: 'gitea',
endpoint: process.env.RENOVATE_ENDPOINT,
enabledManagers: ['docker-compose', 'kubernetes', 'helm-values'],
'helm-values': {
managerFilePatterns: ['/k8s/.+values\\.ya?ml$/'],
},
enabledManagers: ['docker-compose', 'kubernetes'],
kubernetes: {
managerFilePatterns: ['/k8s/.+\\.ya?ml$/'],
},
@@ -25,10 +22,7 @@ data:
dependencyDashboard: true,
prCreation: 'immediate',
labels: ['dependencies', 'automated'],
extends: [
'config:recommended',
':dependencyDashboard',
],
extends: ['config:recommended', ':dependencyDashboard'],
packageRules: [
{
description: 'Do not update private homelab images',
+1 -1
View File
@@ -16,7 +16,7 @@ spec:
restartPolicy: Never
containers:
- name: renovate
image: renovate/renovate:44.103.0
image: renovate/renovate:44.102.0
env:
- name: RENOVATE_PLATFORM
value: gitea
+1 -1
View File
@@ -1,6 +1,6 @@
services:
renovate:
image: renovate/renovate:44.97.2
image: renovate/renovate:44.83.2
container_name: renovate
restart: "no"
env_file: