secubox/secubox-deb
1
0
Fork 0
mirror of https://github.com/CyberMind-FR/secubox-deb.git synced 2026-07-29 17:37:13 +00:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

fix(health-doctor): drop act-runner-arm64 from vital REGISTRY (closes #214) (#215)

Browse Source 
A stopped CI runner doesnt break the board — its an optional capability,
not a vital service. Counting it toward the failing tally noise-floors
the dashboard. Drop from REGISTRY; keep the function for the future
REGISTRY_INFORMATIONAL tier.

Operator feedback after #213 first deploy: 4/11 failing on a healthy
board because act-runner-ci-arm64 LXC was stopped (intentionally), and
two others (gitea/mail LXC) genuinely needed lxc-start. The runner
shouldnt have been in the count.

Co-authored-by: CyberMind-FR <gandalf@Gk2.net>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
CyberMind 2026-05-19 07:55:16 +02:00 committed by GitHub
parent 4736cb32a2
commit b5c8262c17
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 25 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
packages/secubox-health-doctor/api/checks.py
View File

@ -174,16 +174,22 @@ def check_act_runner_arm64() -> tuple[bool, dict]:
# ── Registry ──────────────────────────────────────────────────────────────
# Vital — every entry here counts toward the "failing" tally in the doctor
# summary. Keep this list strict: only services whose absence makes the
# board itself unusable. CI runners, optional capacity probes, etc. should
# go in REGISTRY_INFORMATIONAL once that tier is forged (#214 followup).
REGISTRY: Dict[str, CheckFn] = {
"haproxy": check_haproxy,
"nginx": check_nginx,
"secubox-metrics": check_secubox_metrics,
"secubox-hub": check_secubox_hub,
"mitmproxy-lxc": check_mitmproxy_lxc,
"gitea-lxc": check_gitea_lxc,
"mail-lxc": check_mail_lxc,
"haproxy": check_haproxy,
"nginx": check_nginx,
"secubox-metrics": check_secubox_metrics,
"secubox-hub": check_secubox_hub,
"mitmproxy-lxc": check_mitmproxy_lxc,
"gitea-lxc": check_gitea_lxc,
"mail-lxc": check_mail_lxc,
"cookie-audit-ledger": check_cookie_audit_ledger,
"crowdsec": check_crowdsec,
"filesystems": check_filesystems,
"act-runner-arm64": check_act_runner_arm64,
"crowdsec": check_crowdsec,
"filesystems": check_filesystems,
}
# check_act_runner_arm64 kept as a module-level function (cheap) so a
# future REGISTRY_INFORMATIONAL tier can reuse it without duplication.
# Not vital — a stopped CI runner doesn't break the board (#214).

9
packages/secubox-health-doctor/debian/changelog
View File

@ -1,3 +1,12 @@
secubox-health-doctor (1.0.1-1~bookworm1) bookworm; urgency=medium
* Drop act-runner-arm64 from vital REGISTRY (#214). A stopped CI runner
doesn't break the board and shouldn't count as a vital-service failure.
The check_act_runner_arm64 function is kept at module level for reuse
by a future REGISTRY_INFORMATIONAL tier.
-- Gerald KERMA <devel@cybermind.fr> Tue, 19 May 2026 07:53:56 +0200
secubox-health-doctor (1.0.0-1~bookworm1) bookworm; urgency=medium
* Initial release (issue #212).