• Joined on 2026-03-08
ltms commented on issue fleet/fleetd#421 2026-09-10 07:07:35 +02:00
A lead can neither read nor safely keep its own held peer messages — the only drain is an injectable pane

Correction from the lead who filed this: the cause paragraph is wrong, and the ticket is more important than it says

The original body blames unreadability on a delivery failure:

So the…

ltms closed issue fleet/fleetd#415 2026-09-10 06:57:46 +02:00
The startup coverage line reports "off" for errorPattern, but legacy classification is still running
ltms merged pull request fleet/fleetd#423 2026-09-10 06:57:34 +02:00
fleetd #415: split coverage() feature-state wording by pattern fallback semantics
ltms pushed to main at fleet/fleetd 2026-09-10 06:57:34 +02:00
e60f892efd Merge pull request 'fleetd #415: split coverage() feature-state wording by pattern fallback semantics' (#423) from worker/415-coverage-wording-2cbf9c-5 into main
ce05886831 fleetd #415: pin which UnsetMeaning Fleetd pairs with which pattern key
be123d0ac7 fleetd #415: split coverage() feature-state wording by pattern-key fallback semantics
Compare 3 commits »
ltms commented on issue fleet/fleetd#421 2026-09-10 06:54:54 +02:00
A lead can neither read nor safely keep its own held peer messages — the only drain is an injectable pane

Checked the two claims I made above, so whoever takes this does not have to.

The body really is available at the seam. Confirmed:

// LeadChannel.java
List<LeadMessage> peek();       …
ltms commented on issue fleet/fleetd#417 2026-09-10 06:54:14 +02:00
Sweep for the REVERSE config mismatch: a boot snapshot read where the key is hot

Filed the structural gap as #427 — I said above I wanted to think about what such a checker could actually assert first, so here is the answer.

The concrete thing I did not see when I wrote…

ltms opened issue fleet/fleetd#427 2026-09-10 06:54:03 +02:00
A config key's hot/deferred class is a claim about every consumer, and nothing checks it that way
ltms opened issue fleet/fleetd#426 2026-09-10 06:50:29 +02:00
FleetHealthMonitor.coverage has no test at all, and it feeds fleet_list's healthCoverage field
ltms closed issue fleet/fleetd#417 2026-09-10 06:46:48 +02:00
Sweep for the REVERSE config mismatch: a boot snapshot read where the key is hot
ltms commented on issue fleet/fleetd#417 2026-09-10 06:46:43 +02:00
Sweep for the REVERSE config mismatch: a boot snapshot read where the key is hot

Sweep done. The reverse defect exists, twice. Both filed:

  • #424 — MemberRegistry freezes fleet.architects, so revoking an architect slot does not revoke it. This is the dangerous…
ltms opened issue fleet/fleetd#425 2026-09-10 06:46:21 +02:00
fleet_profiles reports a frozen "default" profile while placement already moved to a new one
ltms opened issue fleet/fleetd#424 2026-09-10 06:45:35 +02:00
Revoking an architect slot does not revoke it: MemberRegistry freezes fleet.architects, and the reload says it applied
ltms opened issue fleet/fleetd#422 2026-09-10 06:39:58 +02:00
Model gating units 2+3: enforce the allow-list at spawn, and turn one model off at runtime without editing profiles
ltms opened issue fleet/fleetd#421 2026-09-10 06:30:17 +02:00
A lead can neither read nor safely keep its own held peer messages — the only drain is an injectable pane
ltms closed issue fleet/fleetd#418 2026-09-10 06:29:43 +02:00
Second instance of the #399 shape: a test uses Phase.ASKING as a barrier for push-loop state published later
ltms merged pull request fleet/fleetd#419 2026-09-10 06:29:23 +02:00
fleetd #418: barrier the throw-path push-loop test on state decide() reads
ltms pushed to main at fleet/fleetd 2026-09-10 06:29:23 +02:00
2d09c8b027 Merge pull request 'fleetd #418: barrier the throw-path push-loop test on state decide() reads' (#419) from worker/418-588283-3 into main
ab0cc71aa4 fleetd #418: barrier the throw-path push-loop test on state decide() reads
Compare 2 commits »
ltms pushed to main at fleet/fleetd 2026-09-10 06:27:38 +02:00
ed54f0224e Merge pull request 'fleetd #416: fleet_list must enumerate the STARTUP profile set' (#420) from worker/416-3ad1da-1 into main
8d5bc3ee89 fleetd #416: fleet_list must enumerate the STARTUP profile set
Compare 2 commits »
ltms merged pull request fleet/fleetd#420 2026-09-10 06:27:37 +02:00
fleetd #416: fleet_list must enumerate the STARTUP profile set
ltms closed issue fleet/fleetd#416 2026-09-10 06:27:37 +02:00
fleet_list advertises free capacity for a profile the spawn gate refuses — CapacitySource reads the live profile keySet