ldraney/pal-e-platform

Fork 0

Close stale pal-e-app / pal-e-docs-app rename tickets after pal-e-production landing #279

New issue

Closed

opened 2026-04-11 20:43:07 +00:00 by forgejo_admin · 3 comments

forgejo_admin commented

2026-04-11 20:43:07 +00:00

Contributor

Type

Bug

Lineage

Discovered during 2026-04-11 routing review. The rename trail pal-e-app → pal-e-docs-app → pal-e-production left a sediment of stale tickets in two directions — some proposing forward renames, some proposing reverse renames. None are accurate. This is housekeeping.

Scope review round 1: NEEDS_REFINEMENT (review-973-2026-04-11). Fixed in this revision: (1) pal-e-app#87 is already closed — treated as "add pointer comment only," not state change, (2) #278 cited explicitly in Related section, (3) AC dropping the feedback_naming_convention lesson-capture (out of scope; that's a memory file, not a pal-e-docs note), (4) AC clarifying that "no new tickets filed by this housekeeping pass" applies to the housekeeping action itself — forward-facing Keycloak / namespace work explicitly DEFERRED under #278.

Repo

forgejo_admin/pal-e-platform

What Broke

Five board / Forgejo items reference repo or namespace names that no longer match the current deployment topology. They confuse new readers, pollute board search, and slow down ticket triage. They are not actionable as written.

Item	State	Title	Why obsolete
`pal-e-platform#234`	open	Fix: pal-e-docs-app pod ImagePullBackOff from Harbor	The `pal-e-docs-app` pod no longer exists. Replaced by `pal-e-production` namespace ~7d ago.
`pal-e-platform#255`	open	Keycloak client rename pal-e-docs-app → pal-e-app	Both names are obsolete. Canonical deployment name is `pal-e-production`.
`pal-e-platform#257`	open	Namespace rename pal-e-docs-app → pal-e-app	Same — namespace is `pal-e-production`.
`pal-e-app#87` (board #510)	already closed (2026-03-28)	Rename pal-e-app → pal-e-docs-app	State already correct. Only needs a pointer comment + board item removal.
`pal-e-app#88` (board #513)	open	Validate: pal-e-app (4 PRs, clone failure)	Repo redirect already happened. Validation target moved.

Repro Steps

Open https://forgejo.tail5b443a.ts.net/forgejo_admin/pal-e-platform/issues/234
Run kubectl get pods -A | grep pal-e-docs-app — observe zero pods
Run kubectl get pods -n pal-e-production — observe the actual frontend pod is here, healthy
Try to clone https://forgejo.tail5b443a.ts.net/forgejo_admin/pal-e-app.git — observe redirect to pal-e-production (or pal-e-docs-app, depending on current state)
Conclude: ticket #234 is unactionable because the resource it references doesn't exist; tickets #255 / #257 reference the wrong target; pal-e-app#88 references a repo that no longer exists at that name; pal-e-app#87 is already closed

Expected Behavior

Each obsolete ticket either:

Closed as wontfix with a comment explaining the rename trail and pointing at forgejo_admin/pal-e-platform#278 (the canonical hostname swap ticket), OR
(For pal-e-app#87 only, which is already closed) gets a pointer comment without a state change

Environment

Cluster: k3s (production)
Forgejo: forgejo.tail5b443a.ts.net
Affected board: board-pal-e-docs (board items #510, #513 reference these issues)
Source of truth for current state: kubectl get pods -A and kubectl get ingress -A as of 2026-04-11
Ticket states confirmed via Forgejo API as of 2026-04-11 (round 1 review-973-2026-04-11)

Per-Ticket Action

pal-e-platform#234 (Fix: pal-e-docs-app ImagePullBackOff) — currently open

Action: close as wontfix. Add closing comment naming the obsolete resource (the pal-e-docs-app pod) and pointing at forgejo_admin/pal-e-platform#278 as the forward path.

pal-e-platform#255 (Keycloak client rename pal-e-docs-app → pal-e-app) — currently open

Action: close as wontfix. Add closing comment explaining that both names in the title are obsolete (canonical is pal-e-production), and that any actual Keycloak client work should be re-scoped under forgejo_admin/pal-e-platform#278 if it's still relevant against the new topology. Do not file a new Keycloak rename ticket from this housekeeping pass — that decision belongs to whoever executes #278.

pal-e-platform#257 (Namespace rename pal-e-docs-app → pal-e-app) — currently open

Action: close as wontfix. Same closing comment pattern. Same deferral note.

pal-e-app#87 (Rename pal-e-app → pal-e-docs-app) — already closed 2026-03-28

Action: NO state change. Add a single pointer comment to the closed ticket noting the rename actually went to pal-e-production and pointing at forgejo_admin/pal-e-platform#278 for the canonical forward path.
Also: remove board item #510 from board-pal-e-docs.

pal-e-app#88 (Validate: pal-e-app) — currently open

Action: close as wontfix. Add closing comment explaining the validation target (pal-e-app repo + 4 PRs) is gone and any new frontend validation should be scoped under #278 against the current pal-e-production deployment.
Also: remove board item #513 from board-pal-e-docs.

Acceptance Criteria

pal-e-platform#234, #255, #257, pal-e-app#88 each closed as wontfix with a comment naming the obsolete resource and citing forgejo_admin/pal-e-platform#278 by full URL
pal-e-app#87 (already closed) receives one pointer comment citing #278 — state unchanged
Board items #510 and #513 are removed from board-pal-e-docs
No new tickets are filed by this housekeeping pass. Forward-facing decisions about Keycloak client rename, namespace rename, or frontend validation are explicitly deferred under #278 — they are NOT this ticket's responsibility and are NOT filed as fresh tickets here. If the executor of #278 decides any of those are still in scope, they'll file new tickets at that point.

Test Expectations

After running: mcp__forgejo__list_issues(owner="forgejo_admin", repo="pal-e-platform", state="open") does NOT contain #234, #255, #257
After running: mcp__forgejo__list_issues(owner="forgejo_admin", repo="pal-e-app", state="open") does NOT contain #88
mcp__pal-e-docs__list_board_items(board_slug="board-pal-e-docs") does NOT contain items 510 or 513
Each closed ticket has a visible closing comment with a working link to #278
pal-e-app#87 has a new pointer comment but its state remains closed (was closed 2026-03-28; this housekeeping pass does not reopen it)

Constraints

Use mcp__forgejo__update_issue to close issues and mcp__forgejo__comment_on_issue to add closing comments
Use mcp__pal-e-docs__remove_board_item to remove board items
Do NOT modify any other tickets, even if they look stale — scope is exactly the 5 listed
Do NOT file new tickets — forward-facing work goes through #278's executor, not this housekeeping pass

Checklist

4 tickets closed as wontfix with closing comments
1 already-closed ticket receives a pointer comment without state change
2 board items removed
Verification queries run and confirm clean state
No unrelated changes

forgejo_admin/pal-e-platform#278 — the canonical hostname swap ticket (the forward path that obsoletes everything in this housekeeping pass)
forgejo_admin/pal-e-api#256 — CORS middleware (the immediate symptom fix that #278 depends on)
arch-domain-pal-e-docs — the current architecture diagram
review-973-2026-04-11 — the round 1 review that caught the original body errors

### Type Bug ### Lineage Discovered during 2026-04-11 routing review. The rename trail `pal-e-app → pal-e-docs-app → pal-e-production` left a sediment of stale tickets in two directions — some proposing forward renames, some proposing reverse renames. None are accurate. This is housekeeping. **Scope review round 1:** NEEDS_REFINEMENT (review-973-2026-04-11). Fixed in this revision: (1) pal-e-app#87 is already closed — treated as "add pointer comment only," not state change, (2) #278 cited explicitly in Related section, (3) AC dropping the `feedback_naming_convention` lesson-capture (out of scope; that's a memory file, not a pal-e-docs note), (4) AC clarifying that "no new tickets filed by this housekeeping pass" applies to the housekeeping action itself — forward-facing Keycloak / namespace work explicitly DEFERRED under #278. ### Repo `forgejo_admin/pal-e-platform` ### What Broke Five board / Forgejo items reference repo or namespace names that no longer match the current deployment topology. They confuse new readers, pollute board search, and slow down ticket triage. They are not actionable as written. | Item | State | Title | Why obsolete | |---|---|---|---| | `pal-e-platform#234` | open | Fix: pal-e-docs-app pod ImagePullBackOff from Harbor | The `pal-e-docs-app` pod no longer exists. Replaced by `pal-e-production` namespace ~7d ago. | | `pal-e-platform#255` | open | Keycloak client rename pal-e-docs-app → pal-e-app | Both names are obsolete. Canonical deployment name is `pal-e-production`. | | `pal-e-platform#257` | open | Namespace rename pal-e-docs-app → pal-e-app | Same — namespace is `pal-e-production`. | | `pal-e-app#87` (board #510) | **already closed (2026-03-28)** | Rename pal-e-app → pal-e-docs-app | State already correct. Only needs a pointer comment + board item removal. | | `pal-e-app#88` (board #513) | open | Validate: pal-e-app (4 PRs, clone failure) | Repo redirect already happened. Validation target moved. | ### Repro Steps 1. Open `https://forgejo.tail5b443a.ts.net/forgejo_admin/pal-e-platform/issues/234` 2. Run `kubectl get pods -A | grep pal-e-docs-app` — observe zero pods 3. Run `kubectl get pods -n pal-e-production` — observe the actual frontend pod is here, healthy 4. Try to clone `https://forgejo.tail5b443a.ts.net/forgejo_admin/pal-e-app.git` — observe redirect to `pal-e-production` (or pal-e-docs-app, depending on current state) 5. Conclude: ticket #234 is unactionable because the resource it references doesn't exist; tickets #255 / #257 reference the wrong target; pal-e-app#88 references a repo that no longer exists at that name; pal-e-app#87 is already closed ### Expected Behavior Each obsolete ticket either: - Closed as `wontfix` with a comment explaining the rename trail and pointing at `forgejo_admin/pal-e-platform#278` (the canonical hostname swap ticket), OR - (For pal-e-app#87 only, which is already closed) gets a pointer comment without a state change ### Environment - Cluster: k3s (production) - Forgejo: `forgejo.tail5b443a.ts.net` - Affected board: `board-pal-e-docs` (board items #510, #513 reference these issues) - Source of truth for current state: `kubectl get pods -A` and `kubectl get ingress -A` as of 2026-04-11 - Ticket states confirmed via Forgejo API as of 2026-04-11 (round 1 review-973-2026-04-11) ### Per-Ticket Action **`pal-e-platform#234`** (Fix: pal-e-docs-app ImagePullBackOff) — currently open - Action: close as `wontfix`. Add closing comment naming the obsolete resource (the `pal-e-docs-app` pod) and pointing at `forgejo_admin/pal-e-platform#278` as the forward path. **`pal-e-platform#255`** (Keycloak client rename pal-e-docs-app → pal-e-app) — currently open - Action: close as `wontfix`. Add closing comment explaining that both names in the title are obsolete (canonical is `pal-e-production`), and that any actual Keycloak client work should be re-scoped under `forgejo_admin/pal-e-platform#278` if it's still relevant against the new topology. **Do not file a new Keycloak rename ticket from this housekeeping pass** — that decision belongs to whoever executes #278. **`pal-e-platform#257`** (Namespace rename pal-e-docs-app → pal-e-app) — currently open - Action: close as `wontfix`. Same closing comment pattern. Same deferral note. **`pal-e-app#87`** (Rename pal-e-app → pal-e-docs-app) — **already closed 2026-03-28** - Action: NO state change. Add a single pointer comment to the closed ticket noting the rename actually went to `pal-e-production` and pointing at `forgejo_admin/pal-e-platform#278` for the canonical forward path. - Also: remove board item #510 from `board-pal-e-docs`. **`pal-e-app#88`** (Validate: pal-e-app) — currently open - Action: close as `wontfix`. Add closing comment explaining the validation target (pal-e-app repo + 4 PRs) is gone and any new frontend validation should be scoped under #278 against the current `pal-e-production` deployment. - Also: remove board item #513 from `board-pal-e-docs`. ### Acceptance Criteria - [ ] `pal-e-platform#234`, `#255`, `#257`, `pal-e-app#88` each closed as `wontfix` with a comment naming the obsolete resource and citing `forgejo_admin/pal-e-platform#278` by full URL - [ ] `pal-e-app#87` (already closed) receives one pointer comment citing #278 — **state unchanged** - [ ] Board items #510 and #513 are removed from `board-pal-e-docs` - [ ] No new tickets are filed by this housekeeping pass. Forward-facing decisions about Keycloak client rename, namespace rename, or frontend validation are explicitly **deferred under #278** — they are NOT this ticket's responsibility and are NOT filed as fresh tickets here. If the executor of #278 decides any of those are still in scope, they'll file new tickets at that point. ### Test Expectations - [ ] After running: `mcp__forgejo__list_issues(owner="forgejo_admin", repo="pal-e-platform", state="open")` does NOT contain #234, #255, #257 - [ ] After running: `mcp__forgejo__list_issues(owner="forgejo_admin", repo="pal-e-app", state="open")` does NOT contain #88 - [ ] `mcp__pal-e-docs__list_board_items(board_slug="board-pal-e-docs")` does NOT contain items 510 or 513 - [ ] Each closed ticket has a visible closing comment with a working link to #278 - [ ] pal-e-app#87 has a new pointer comment but its state remains `closed` (was closed 2026-03-28; this housekeeping pass does not reopen it) ### Constraints - Use `mcp__forgejo__update_issue` to close issues and `mcp__forgejo__comment_on_issue` to add closing comments - Use `mcp__pal-e-docs__remove_board_item` to remove board items - Do NOT modify any other tickets, even if they look stale — scope is exactly the 5 listed - Do NOT file new tickets — forward-facing work goes through #278's executor, not this housekeeping pass ### Checklist - [ ] 4 tickets closed as wontfix with closing comments - [ ] 1 already-closed ticket receives a pointer comment without state change - [ ] 2 board items removed - [ ] Verification queries run and confirm clean state - [ ] No unrelated changes ### Related - `forgejo_admin/pal-e-platform#278` — the canonical hostname swap ticket (the forward path that obsoletes everything in this housekeeping pass) - `forgejo_admin/pal-e-api#256` — CORS middleware (the immediate symptom fix that #278 depends on) - `arch-domain-pal-e-docs` — the current architecture diagram - `review-973-2026-04-11` — the round 1 review that caught the original body errors

forgejo_admin commented

2026-04-11 20:57:21 +00:00

Author

Contributor

Scope Review: NEEDS_REFINEMENT

Review note: review-973-2026-04-11

Housekeeping scope is directionally sound. One of the 5 listed tickets is already closed, and there are a few small body accuracy issues to resolve before this can advance to todo. No decomposition needed — single-agent close-out, ~2-3 min of API calls.

Must-fix before todo:

[BODY] pal-e-app#87 is already closed (2026-03-28T17:01:43Z). Acknowledge the existing closed state in the decision table; only a pointer comment + board item #510 removal is needed, not a state change.
[BODY] Related section says "the canonical hostname swap ticket" without a number. Cite forgejo_admin/pal-e-platform#278 explicitly so closing comments can link it unambiguously.
[BODY] AC #5 references feedback_naming_convention — this is a local memory filename, not a confirmed pal-e-docs note slug. Recommendation: drop this AC; lesson-capture is out of scope for a closure pass and grows it into a two-agent job.
[SCOPE] Reconcile the tension between AC #4 ("no new tickets filed by this pass") and the per-ticket decision notes for #255/#257 that say "separate ticket if anyone cares to file it." Explicitly defer any Keycloak/namespace rename-to-current-topology intent as follow-up under #278.

Verified:

Board items #510 and #513 exist on board-pal-e-docs, ready to remove
#234, #255, #257, pal-e-app#88 all confirmed open at the Forgejo API
story:superuser-maintain verified in project-pal-e-docs user-stories section
Depends on #278 (hostname swap) — respect spawn-prompt guidance to land after the dust settles

Flag for Ava: If the Keycloak client ID currently drifts from the pal-e-production deployment name, a forward-facing Keycloak rename ticket should be queued under #278 — separate decision.

## Scope Review: NEEDS_REFINEMENT Review note: `review-973-2026-04-11` Housekeeping scope is directionally sound. One of the 5 listed tickets is already closed, and there are a few small body accuracy issues to resolve before this can advance to `todo`. No decomposition needed — single-agent close-out, ~2-3 min of API calls. **Must-fix before todo:** - `[BODY]` **pal-e-app#87 is already closed** (2026-03-28T17:01:43Z). Acknowledge the existing closed state in the decision table; only a pointer comment + board item #510 removal is needed, not a state change. - `[BODY]` Related section says "the canonical hostname swap ticket" without a number. Cite `forgejo_admin/pal-e-platform#278` explicitly so closing comments can link it unambiguously. - `[BODY]` AC #5 references `feedback_naming_convention` — this is a local memory filename, not a confirmed pal-e-docs note slug. Recommendation: drop this AC; lesson-capture is out of scope for a closure pass and grows it into a two-agent job. - `[SCOPE]` Reconcile the tension between AC #4 ("no new tickets filed by this pass") and the per-ticket decision notes for #255/#257 that say "separate ticket if anyone cares to file it." Explicitly defer any Keycloak/namespace rename-to-current-topology intent as follow-up under #278. **Verified:** - Board items #510 and #513 exist on board-pal-e-docs, ready to remove - #234, #255, #257, pal-e-app#88 all confirmed open at the Forgejo API - `story:superuser-maintain` verified in project-pal-e-docs user-stories section - Depends on #278 (hostname swap) — respect spawn-prompt guidance to land after the dust settles **Flag for Ava:** If the Keycloak client ID currently drifts from the `pal-e-production` deployment name, a forward-facing Keycloak rename ticket should be queued under #278 — separate decision.

forgejo_admin referenced this issue

2026-04-11 20:58:38 +00:00

Hostname swap step 1 — add additive api.pal-e-docs funnel via pal-e-deployments kustomize #278

forgejo_admin commented

2026-04-11 21:01:59 +00:00

Author

Contributor

Scope Review r2: APPROVED

Review note: review-973-2026-04-11-r2

All four round 1 [BODY] findings are resolved in this revision:

pal-e-app#87 pointer-comment-only — What Broke table, Per-Ticket Action, AC#2, and Test Expectations all correctly treat #87 as already-closed (2026-03-28) with no state change.
#278 cited by full reference — Related section lists forgejo_admin/pal-e-platform#278 explicitly; Per-Ticket Action references #278 throughout.
feedback_naming_convention AC dropped — AC count now 4; Lineage explicitly documents the drop as out-of-scope.
"No new tickets" reconciled — Per-Ticket Action for #255/#257 and AC#4 explicitly defer forward-facing Keycloak/namespace/validation work under #278. "If anyone cares" language removed.

Scope fits a single agent pass (~2–3 min: 4 close+wontfix, 5 comments, 2 board removals). No decomposition needed. Blast radius low, all operations reversible.

Board item #973 can advance backlog → todo.

Execution-order reminder (not a review gate): do not dispatch dev work on #973 until #278 is at least in_progress, per original spawn guidance.

## Scope Review r2: APPROVED Review note: `review-973-2026-04-11-r2` All four round 1 [BODY] findings are resolved in this revision: 1. **pal-e-app#87 pointer-comment-only** — What Broke table, Per-Ticket Action, AC#2, and Test Expectations all correctly treat #87 as already-closed (2026-03-28) with no state change. 2. **#278 cited by full reference** — Related section lists `forgejo_admin/pal-e-platform#278` explicitly; Per-Ticket Action references #278 throughout. 3. **feedback_naming_convention AC dropped** — AC count now 4; Lineage explicitly documents the drop as out-of-scope. 4. **"No new tickets" reconciled** — Per-Ticket Action for #255/#257 and AC#4 explicitly defer forward-facing Keycloak/namespace/validation work under #278. "If anyone cares" language removed. Scope fits a single agent pass (~2–3 min: 4 close+wontfix, 5 comments, 2 board removals). No decomposition needed. Blast radius low, all operations reversible. Board item #973 can advance `backlog → todo`. **Execution-order reminder (not a review gate):** do not dispatch dev work on #973 until #278 is at least `in_progress`, per original spawn guidance.

forgejo_admin referenced this issue

2026-04-11 21:04:12 +00:00

Hostname swap follow-up — Steps 2-7 roadmap (after additive funnel lands) #280

forgejo_admin commented

2026-04-12 15:38:14 +00:00

Author

Contributor

All actions complete:

pal-e-platform#234 — closed as wontfix with comment
pal-e-platform#255 — closed as wontfix with comment
pal-e-platform#257 — closed as wontfix with comment
pal-e-production#87 — pointer comment added (state unchanged)
pal-e-production#88 — closed as wontfix with comment
board-pal-e-docs items #510, #513 — removed from board

All actions complete: 1. **pal-e-platform#234** — closed as wontfix with comment 2. **pal-e-platform#255** — closed as wontfix with comment 3. **pal-e-platform#257** — closed as wontfix with comment 4. **pal-e-production#87** — pointer comment added (state unchanged) 5. **pal-e-production#88** — closed as wontfix with comment 6. **board-pal-e-docs items #510, #513** — removed from board

forgejo_admin closed this issue

2026-04-12 15:38:14 +00:00