Implement rotating JSONL recorder for persistent logging#10428
Merged
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
Implements an always-on, process-global persistent recorder in the MCP server that captures Meshtastic pubsub events into rotating JSONL streams under mcp-server/.mtlog/, and exposes read/query tools (logs/telemetry/packets/events windows, export, pause/resume) plus optional build-time macro injection for leak-hunting workflows.
Changes:
- Add recorder write-side (rotating JSONL writer, log/telemetry/packet parsing, pubsub subscriptions) and read-side query helpers (
log_query.py) with new MCP tools inserver.py. - Add
build_flagssupport forflash.build()/flash.flash()propagated viaPLATFORMIO_BUILD_FLAGSandpio.run(extra_env=...). - Add unit tests for recorder/query behavior and build-flag propagation, plus a Datadog forwarder + dashboard JSON and updated agent commands.
Reviewed changes
Copilot reviewed 17 out of 17 changed files in this pull request and generated 9 comments.
Show a summary per file
| File | Description |
|---|---|
| mcp-server/tests/unit/test_recorder.py | New unit tests covering recorder parsing, heap synthesis, rotation, and query APIs. |
| mcp-server/tests/unit/test_build_flags.py | New unit tests for build_flags → PLATFORMIO_BUILD_FLAGS translation and propagation to pio.run. |
| mcp-server/src/meshtastic_mcp/server.py | Starts recorder on import and adds MCP tools for querying/exporting recorder streams; adds build_flags args to build/flash tools. |
| mcp-server/src/meshtastic_mcp/serial_session.py | Publishes serial monitor lines to pubsub so the recorder can capture text-mode output. |
| mcp-server/src/meshtastic_mcp/recorder/rotating.py | New size-capped rotating JSONL writer with gzip archives + pruning. |
| mcp-server/src/meshtastic_mcp/recorder/recorder.py | New process-global Recorder singleton that subscribes to pubsub topics and writes logs/telemetry/packets/events streams. |
| mcp-server/src/meshtastic_mcp/recorder/parsers.py | New best-effort parsing for firmware log line prefixes and telemetry variants. |
| mcp-server/src/meshtastic_mcp/recorder/init.py | New package entrypoint exporting Recorder / get_recorder. |
| mcp-server/src/meshtastic_mcp/pio.py | Adds extra_env support to subprocess execution for build-time env injection. |
| mcp-server/src/meshtastic_mcp/log_query.py | New streaming queries over recorder JSONL streams (windows, telemetry downsampling/slope, export). |
| mcp-server/src/meshtastic_mcp/flash.py | Adds build_flags support to build/flash and threads flags through to pio.run(extra_env=...). |
| mcp-server/scripts/mtlog_to_datadog.py | New forwarder shipping .mtlog logs/telemetry to Datadog with cursor-based resume. |
| mcp-server/scripts/datadog-dashboard.json | New sample Datadog dashboard definition for recorder-derived metrics/logs. |
| mcp-server/.gitignore | Ignores .mtlog/ recorder output and cursor file. |
| .claude/commands/repro.md | Updates repro workflow to export per-attempt recorder slices. |
| .claude/commands/leakhunt.md | New leakhunt workflow documentation leveraging recorder tools and heap slope. |
| .claude/commands/diagnose.md | Updates diagnose workflow to include recorder-based error + heap trend checks. |
Contributor
Author
|
@copilot apply changes based on the comments in this thread |
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/2541773c-869a-463f-9fae-8505272c06ff Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Contributor
Applied the remaining review-thread fixes in 13d0a16: archive pruning now follows filename chronology, |
The previous "Fixes" commit added `_files_snapshot()` which acquires
`self._lock` so handlers don't race with `stop()` clearing `_files`.
But two callers were already holding `self._lock` when they invoked
methods that go through the snapshot:
- `start()` writes the `recorder_start` event from inside its `with
self._lock:` block. `_write_event` -> `_files_snapshot` re-acquires
the same non-reentrant `threading.Lock`, freezing process startup.
- `force_rotate_all()` calls `self.status()` (which also acquires
`self._lock`) while still holding the lock from rotating each file.
Both fixes release the lock before the call. The recorder_start marker
still lands in events.jsonl because the started/started_at flags are
already set when we write it.
Verified end-to-end against the standalone /tmp/verify_pr_fixes.py
harness — all 9 PR review-comment fixes pass, including pause/resume
event ordering and concurrent start/stop without KeyError.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Contributor
Author
|
@copilot apply changes based on the comments in this thread |
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com>
Contributor
Applied the latest review-thread fixes in 7714338: recorder startup is now best-effort in |
balya
pushed a commit
to balya/meshtastic-firmware
that referenced
this pull request
May 14, 2026
…10428) * Implement rotating JSONL recorder for persistent logging * Fixes * Update documentation and clean up imports in command files * Address remaining recorder review feedback Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/2541773c-869a-463f-9fae-8505272c06ff Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * recorder: fix lock re-entry deadlock on start() and force_rotate_all() The previous "Fixes" commit added `_files_snapshot()` which acquires `self._lock` so handlers don't race with `stop()` clearing `_files`. But two callers were already holding `self._lock` when they invoked methods that go through the snapshot: - `start()` writes the `recorder_start` event from inside its `with self._lock:` block. `_write_event` -> `_files_snapshot` re-acquires the same non-reentrant `threading.Lock`, freezing process startup. - `force_rotate_all()` calls `self.status()` (which also acquires `self._lock`) while still holding the lock from rotating each file. Both fixes release the lock before the call. The recorder_start marker still lands in events.jsonl because the started/started_at flags are already set when we write it. Verified end-to-end against the standalone /tmp/verify_pr_fixes.py harness — all 9 PR review-comment fixes pass, including pause/resume event ordering and concurrent start/stop without KeyError. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * Fix markdown linting issues in leakhunt.md and repro.md * Handle recorder startup and query review fixes Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Tighten recorder follow-up tests Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Stabilize recorder startup tests Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Remove brittle recorder startup test Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Polish recorder follow-up errors Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Refine recorder startup and regex errors Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Clean up recorder follow-up nits Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Trunk --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Evil8it
pushed a commit
to Evil8it/ME4TACTNK
that referenced
this pull request
Jun 10, 2026
…10428) * Implement rotating JSONL recorder for persistent logging * Fixes * Update documentation and clean up imports in command files * Address remaining recorder review feedback Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/2541773c-869a-463f-9fae-8505272c06ff Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * recorder: fix lock re-entry deadlock on start() and force_rotate_all() The previous "Fixes" commit added `_files_snapshot()` which acquires `self._lock` so handlers don't race with `stop()` clearing `_files`. But two callers were already holding `self._lock` when they invoked methods that go through the snapshot: - `start()` writes the `recorder_start` event from inside its `with self._lock:` block. `_write_event` -> `_files_snapshot` re-acquires the same non-reentrant `threading.Lock`, freezing process startup. - `force_rotate_all()` calls `self.status()` (which also acquires `self._lock`) while still holding the lock from rotating each file. Both fixes release the lock before the call. The recorder_start marker still lands in events.jsonl because the started/started_at flags are already set when we write it. Verified end-to-end against the standalone /tmp/verify_pr_fixes.py harness — all 9 PR review-comment fixes pass, including pause/resume event ordering and concurrent start/stop without KeyError. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * Fix markdown linting issues in leakhunt.md and repro.md * Handle recorder startup and query review fixes Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Tighten recorder follow-up tests Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Stabilize recorder startup tests Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Remove brittle recorder startup test Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Polish recorder follow-up errors Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Refine recorder startup and regex errors Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Clean up recorder follow-up nits Agent-Logs-Url: https://github.com/meshtastic/firmware/sessions/78540a9f-fe62-4350-b252-0ae5621f0b8a Co-authored-by: thebentern <9000580+thebentern@users.noreply.github.com> * Trunk --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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
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.
No description provided.