Move L3 enforcement to the host firewall-bouncer (systemd, nftables): drop the Traefik plugin, its secrets volume and the crowdsec Middleware, remove bouncer refs from all IngressRoutes. Disable the http-generic-bf scenario (403-burst bans hurt legit automation under L3 enforcement). Add a Gateway API PoC for homepages prod and CrowdSec PrometheusRule alerts.
RollingUpdate with default maxSurge needs a spare pod the single node does not have (99% CPU requested), so multi-workload restarts end Pending and verify times out. Recreate on all replicas:1 Deployments (immich-server and bentopdf keep RollingUpdate at replicas 2). Also trims CPU/memory requests toward measured use (adguard, authentik, gitea, netbox, uptime-kuma, netbird-server) and gives traefik requests/limits so it is no longer BestEffort.
Finishes the sizing pass over every workload the deploy actually manages. Each
request is at or above the container's p95 over the last seven days, so nothing
is sized below what it is known to use, and each limit is between 1.6x and 5x
the observed max, which is the figure that decides whether a burst gets an
OOMKill.
Some of these go up, and that is the point. adguard was holding 975M against a
500Mi request and netbox 962M against 512Mi, so both sat permanently above
their own request and were standing eviction candidates on a node that has
about 300M of headroom. Raising a request costs scheduler room; leaving it low
costs the pod its place in the queue when the node gets tight.
Others come down. loki ran with a 2Gi limit on 224M, gitea 1.5Gi on 305M, the
authentik worker 1Gi on 305M, and a tail of single-purpose pods -- redis,
glance, the two homepages, cfddns, session-keeper, the netbird dashboard, the
loki gateway -- each reserved 4x to 16x more than they have ever touched.
prometheus gets the opposite treatment: 768Mi/2560Mi, above its p95, because it
compacts its TSDB in place and that is a burst worth budgeting for rather than
throttling.
Two of these limits are close enough to the observed max to be worth watching
rather than trusting: adguard at 1.5x, and its DNS cache grows monotonically, so
the ceiling is a date, not a margin. That was true before this change too; the
pod sizing does not fix it and the cache needs bounding.
CPU limits are untouched throughout. Leaving postgres alone as well: it sits in
an uncommitted file that belongs to other work in progress.
Verified: every request is at or above p95 and every limit above the observed
max across all 74 containers, and 16/16 local gates pass.
ruff.toml has declared `quote-style = "single"` and line-length 120 since the
lint job landed, and 118 of 128 files follow it. The panel backend and the
netbox configuration were written in black/prettier style instead, so a
`ruff format --check` would have failed on them from the start.
Bring them onto the style the repository already declares, which is what makes
the check adoptable at all. Formatting only: apart from quote style the diff is
multi-line expressions joined where they fit inside 120 columns.
Both suites still pass afterwards (25 pytest, and ruff check is clean).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
python -c 'exec /usr/bin/curl ...' is shell syntax, not Python, so every probe raised SyntaxError and the pod never became ready. Verified curl against /login/ returns HTTP 200.
compose.yaml is a local-only stand on 127.0.0.1:8000 per README; the live service runs in-cluster. The root active marker made apply-compose fail on gitignored .env vars.
forust
