fix: harden memory and kanban reliability - #31564
Open
Thawpoint wants to merge 2 commits into
Open
Conversation
- add Hindsight memory setup/config validation coverage - harden Kanban notifier cursors, reply targets, and dedicated notification bot routing - add Kanban DB integrity/repair commands with cooperative maintenance locking - add watchdog/runbook coverage for low-noise Kanban cron monitoring
alt-glitch
added
type/bug
teknium1
reviewed
Jul 13, 2026
teknium1
left a comment
teknium1
left a comment
Contributor
There was a problem hiding this comment.
Thanks for the broad reliability work. Some of the targeted failure modes are already addressed on current main, but two parts need rework before any salvage.
Problems
gateway/run.py:4673changes notifier delivery to("completed",)only. Current main deliberately preserves abnormal-event notification and retry-cycle delivery ingateway/kanban_watchers.py:165-179;tests/hermes_cli/test_kanban_notify.py:82-149covers this contract.gateway/run.py:4649introducesHERMES_KANBAN_NOTIFY_IN_GATEWAYdespite addingkanban.notify_in_gatewayin the same patch.AGENTS.md:102-107requires behavioral settings to useconfig.yaml, not new non-secretHERMES_*variables.
Suggested changes
- Preserve the existing abnormal-event notifier contract; isolate any completion-only policy as a separately configured and tested change.
- Use only
kanban.notify_in_gatewayfor this behavior.
Automated hermes-sweeper review.
| # task is genuinely done lets the cursor (advanced atomically by | ||
| # claim_unseen_events_for_sub) handle dedup, and any retry-loop | ||
| # event reaches the user. | ||
| TERMINAL_KINDS = ("completed",) |
Contributor
There was a problem hiding this comment.
This suppresses blocked/gave_up/crashed/timed_out notifications. Current main intentionally keeps those events visible and retains subscriptions across retry cycles (gateway/kanban_watchers.py:165-179; tests/hermes_cli/test_kanban_notify.py:82-149). Please preserve that contract or split a completion-only policy into a separately configured change.
| except Exception: | ||
| logger.warning("kanban notifier: config loader unavailable; disabled") | ||
| return | ||
| env_override = os.environ.get("HERMES_KANBAN_NOTIFY_IN_GATEWAY", "").strip().lower() |
Contributor
There was a problem hiding this comment.
Please remove this new non-secret behavioral environment override. The patch already adds kanban.notify_in_gateway; AGENTS.md:102-107 requires user-facing behavioral settings to live in config.yaml rather than new HERMES_* variables.
teknium1
added
sweeper:risk-session-state
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
4 participants
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.
Summary
Test Plan
python -m pytest -o 'addopts=' tests/hermes_cli/test_kanban_blocked_sticky.py tests/hermes_cli/test_kanban_db.py::test_recompute_ready_does_not_immediately_reopen_gave_up_tasks -qpython -m pytest -o 'addopts=' tests/cron/test_scheduler.py tests/cron/test_vicky_kanban_cron_watchdogs.py tests/gateway/test_kanban_notifier.py tests/gateway/test_kanban_reply_targets.py tests/gateway/test_restart_resume_pending.py tests/hermes_cli/test_config.py tests/hermes_cli/test_kanban_blocked_sticky.py tests/hermes_cli/test_kanban_core_functionality.py tests/hermes_cli/test_kanban_db.py tests/hermes_cli/test_kanban_notify.py tests/hermes_cli/test_kanban_specify.py tests/plugins/memory/test_hindsight_provider.py tests/test_hermes_state_wal_fallback.py tests/tools/test_kanban_tools.py -q