fix(release): don't false-abort flash on a resumed subscription; tidy #12 console comments - #93
Merged
Merged
Conversation
…12 console comments check_subscription_log (both ota-release.sh and esp32-release.sh) died when the matter-server log had no fresh '(Re-)Subscription succeeded' line in the last 15m. But the flash gate already re-interviews the node and polls it back to available, and matter-server only marks a node available once its subscription is up -- so availability already asserts the subscription (#64). matter-server sometimes RESUMES a subscription after the re-interview without logging a new line (seen on the 2026-07-23 reject flip), which false-aborted a healthy flash mid-run. Warn instead of die when the line is absent; the primary availability gate still holds. Also refresh the #12 checksum-mismatch counter comments in both diag consoles (esp32 diag_console.cpp + ameba hisense_diag_console.h): the RX checksum verify now REJECTS a mismatch (skips parse + counts a link-miss), so the tally is the count of corrupt 0x66 frames dropped, not the old log-only observation. Bumps firmware/src/version.txt 1.3.23 -> 1.3.24 (ameba console comment lives in firmware/src; CI requires the int to increase). Comment-only for the firmware -- nodes 14/62 stay on 1.3.23, no reflash needed. Assisted-by: AI
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
The 6am reject-flip run false-aborted at
check_subscription_log: matter-server had no fresh(Re-)Subscription succeededline in the last 15m, so the gatedie'd even though the node was healthy. matter-server sometimes resumes a subscription after the flash's re-interview without logging a new line.What
check_subscription_log(bothota-release.sh+esp32-release.sh): warn instead ofdiewhen the log line is absent. The flash gate already re-interviews the node and polls it back to available, and matter-server only marks a node available once its subscription is up — so availability already asserts the subscription (QA gap: nothing exercises a Matter subscription, so a broken data model passes every gate #64). A missing log line is not proof of a break; a falsedieaborts a healthy flash mid-run.esp32-matter/main/diag_console.cpp+src/sdk-edits/hisense_diag_console.h): the RX checksum verify now rejects a mismatch (skips parse + counts a link-miss), so the mismatch tally counts corrupt0x66frames dropped — not the old log-only observation.firmware/src, so CI requires the int to increase. Comment-only — nodes 14/62 stay on 1.3.23, no reflash.Field state
All three nodes current: node 35 (ESP32) 1.1.7, node 14 (kitchen ameba) 1.3.23, node 62 (office ameba) 1.3.23 — all on the reject build, 0 checksum mismatches.