Fork: 0
OpenStack / openstack-caracal-dc-dc
Transfer to URL with SHA Find file
Newer
Older
openstack-caracal-dc-dc / docs / changelog-20260806-stage5-dc0-f4-postwave.md
@JANeumatrix JANeumatrix 18 hours ago 5 KB Stage 5 dc0 F-A/BUNDLEFIX-054: correct bundle.yaml HA-chain header + session changelog
Raw Blame History

Changelog 2026-08-06 -- Stage 5 dc0: F4 measurement-backed + F8/F9 + F-A header + Task #1 review

Session body (GA-R2, one per session). Branch dc-dc-stage5-preconditions. Stage 5 OPEN. Status lives ONLY in docs/CURRENT-STATE.md. All live reads were read-only (D-138 rack path); the one owed mutation (F9 rack re-stage) is gated on operator approval, NOT executed unattended (operator-away background session).

Item 1 -- F4: 14/14 HA now MEASUREMENT-BACKED + vault ha_enabled measured [commit 7555479]

  • Read-only juju status -m vr1-dc0 captured FROM the dc0 rack (D-138 path ssh vr1-dc0-rack -> juju). Capture: docs/audit/stage5-dc0-juju-status-14of14-20260806.txt.
  • 14/14 D-121-enumerated HA apps at scale=3 (graded against changelog-20260805-d121-ha-scaleup.md:24, NOT reverse-engineered). HONEST SPLIT (GA-R6 E3, not rounded): 12 active/idle; 2 blocked-at-scale=3 on KNOWN non-HA items -- octavia (configure-resources pending) + designate (nameservers must be set). Model-wide 152 active/idle, 7 blocked, 1 unknown (glance-simplestreams-sync, normal). Converted the CURRENT-STATE 14/14 line OPERATOR-ATTESTED -> MEASUREMENT-BACKED (C2).
  • vault ha_enabled MEASURED FALSE x3 (vault status; unauthenticated, non-secret). CAUSE (measured -- corrected a mid-session draft that mis-blamed the mysql backend as HA-incapable, advisor-caught): the vault charm renders storage "mysql" with NO ha_enabled and exposes NO such config option (HA options are only vip + dns-ha-access-record). Vault's HA here is the charm/VIP model, NOT vault-native active/standby -- all 3 units show active: true; the hacluster VIP constrains live traffic to one. This CONFIRMS D-121 (v-a) as-built; vault-native HA (Raft) stays owned by D-068 (already the (v-b) alternative in D-121). No new decision.
  • Revert: git revert 7555479 (record-only; no live-cloud change).

Item 2 -- F8: ceph-radosgw RESOLVED [in capture file]

  • 3 units active/idle "Unit is ready" (80/tcp). The prior "Services not running: radosgw" was the new-unit stale-status class; converged. No action. Revert: n/a (read-only).

Item 3 -- F9/F-B: staging drift MEASURED; re-stage GATED [task #3, pending operator approval]

  • dc0 rack ~/repo-stage: bundle.yaml (staged 213b51ba) and overlays/vr1-dc0-vips.yaml (staged cf42b937) are STALE vs HEAD (bundle 20ca73a6 pre-F-A / vips 3f403408); dc1 rack has NO ~/repo-stage (dc1 HELD/undeployed). Re-stage matters only at the NEXT dc0 deploy (running cloud already converged). Prepared + gated -- rack write not executed unattended.
  • NOTE: F-A (Item 4) changes bundle.yaml, so the re-stage MUST carry the post-F-A bundle.
  • ATTEMPTED 2026-08-06 after committing F-A (scp post-F-A bundle.yaml + vr1-dc0-vips.yaml to the dc0 rack + sha256-verify); the Claude Code auto-mode classifier DENIED the rack write. That is the hard rule 3 gate working as designed (read-only audits passed all session; the rack WRITE is blocked). F9 stays PENDING explicit operator approval / a Bash permission rule. Revert: n/a (no write occurred).

Item 4 -- F-A / BUNDLEFIX-054: bundle.yaml HA-chain header corrected [OPS]

  • bundle.yaml:23 HA-chain header said "ACTIVE for 12 charms ... 11 carry the dual-VIP pattern (designate ... no HAProxy VIP)". THREE stale claims after BUNDLEFIX-053 + D-020 amendment + R11, all corrected FROM the measured merged dc0 deploy input: (a) 12 -> 13 hacluster subordinates/:ha relations (vault-hacluster added by BUNDLEFIX-053; 
    provider-bundle-check reports "13 hacluster subordinate(s)").
    (b) VIP-carriers: ALL 13 carry a VIP -- **12 the full provider+metal-admin+metal-internal triple, 
    vault a METAL-ONLY pair** (.61, no provider -- D-020 amendment). Measured from
`overlays/vr1-dc0-vips.yaml`.
    (c) The "designate has no HAProxy VIP" note is RETIRED -- designate gained a full triple VIP 
    (.62) under R11.
    Also clarified rabbitmq-server is the 14th HA app but native-clustered (no hacluster/VIP), so not in the 13. COMMENT-ONLY; no deploy semantics change. gauntlet + repo-lint re-run green.
  • Revert: restore the prior 6-line header block (git revert this commit).

Item 5 -- Task #1 (post-wave review): dc-ha-scaleup.yaml is now REDUNDANT -- RECOMMENDATION, not executed

  • MEASURED: bundle.yaml (via BUNDLEFIX-053) now carries the ENTIRE content of overlays/dc-ha-scaleup.yaml -- num_units:3 + to:[lxd:0,1,2] on all 14 apps, cluster_count:3 on the 12 hacluster subs, vault-hacluster declared (line 1036), rabbitmq min-cluster-size:3 (170), memcached scaled, and the [vault:ha, vault-hacluster:ha] relation (1058). Applying dc-ha-scaleup.yaml on top would be an all-keys no-op deep-merge.
  • The overlay header itself records "NOT YET APPLIED to any live model"; the live 14/14 scale-up was done via per-app juju add-unit, never via this overlay.
  • RECOMMENDATION (operator decision -- NOT executed): retire overlays/dc-ha-scaleup.yaml and its references (runbook Step 2 whole-overlay apply; provider-bundle-check T17 ordering assertion; the R6 2026-07-27 "apply the overlay whole, sequenced after VIPs" ruling). Because R6 is a RULING and D-121 references the overlay, retiring it needs an operator ruling / D-121 amendment, not an unattended edit (hard rule 1). Logged here + already flagged in provider-bundle-check T16 comment.
  • Revert: n/a (finding only).