Conversation
This was referenced Sep 25, 2026
DeusData
force-pushed
the
fix/issue-2167
branch
from
September 30, 2026 20:13
886e6da to
6469d3b
Compare
#2167) The background watcher only watches the project an open MCP session is rooted in, and drops the watch when the last session for that project closes. A repository indexed with `cli index_repository` (or from a session rooted elsewhere) is never watched, but README promised "Auto-sync keeps it fresh after that" for every index_repository target and the MCP instructions said "Indexes auto-refresh". Nothing in the status output said which projects were actually watched, so a stale CLI-indexed repo looked like a watcher bug. What gets watched is unchanged. Instead: - index_status gains a `watch` object: `watched`, plus `strategy`, `poll_interval_ms` and `last_scan_at` when watched, or a `reason` (not_session_project, cli_session, auto_watch_off, watcher_disabled, not_registered, no_watcher) and a re-index hint when not. It does not touch the #1561 `freshness` key. - The watcher publishes per-project strategy, cadence and last completed scan time through atomics and exposes them via cbm_watcher_project_info(); the daemon answers index_status through a session-scoped provider. - README, docs/CONFIGURATION.md, docs/index.html, the index_status tool description and the MCP instructions now state the real scope. `watcher_enabled` already shows in `config list` since v0.11.0 (680bc72). Signed-off-by: Martin Vogel <martin.vogel.tech@gmail.com>
DeusData
added a commit
that referenced
this pull request
Sep 30, 2026
…aemon (#2183, #2167) README.md:132, README.md:633, README.md:761, and docs/CONFIGURATION.md:203 said one-shot `cli` commands "never start or connect to the coordination daemon" / read "their own environment without starting the daemon". That was already contradicted a few lines above it (README.md:126 lists one-shot CLI commands among the processes that "share a crash-safe OS admission barrier" with the daemon) and by the code: main_local_cli_daemon_execute() (src/main.c) bootstraps a client connection to the shared per-user coordination daemon via main_client_bootstrap_with_upgrade(), spawning one (bootstrap.daemon_spawned) when none is running, before dispatching the tool call. Verified live: `cli --verbose list_projects` against an isolated HOME/CBM_CACHE_DIR/CBM_RUNTIME_DIR printed "hint: this command started a temporary CBM daemon", and its cbm-daemon.log showed daemon.start -> daemon.runtime_stopping (reason=last_committed_client_disconnected) -> daemon.lifetime_end, i.e. the daemon it spawned exited again once the CLI command's own connection closed. bootstrap_production_spawn() passes the calling process's `environ` straight to posix_spawn(), so a CLI invocation that starts the daemon also seeds that daemon's captured daemon-owned environment (CBM_DIAGNOSTICS, CBM_LOG_LEVEL, etc.) - it does not read "its own environment" independently as the docs claimed. Corrected all four passages to state the real behavior: CLI commands connect to the shared daemon (starting one if none is running) for the same admission barrier and per-project locks, hold a `cli_session` that is never registered with the background watcher, and - if their own connection is what started the daemon - that daemon exits again once the command's connection closes and no other session is attached. No opt-out env var or flag for an in-process/no-daemon CLI mode exists (grepped the whole tree), so none is documented. This is a pure documentation fix; no production behavior changed. Kept clear of PR #2348 (fix/issue-2167), which separately rewrites the watcher-scope paragraph at README.md:152/237/672 and docs/CONFIGURATION.md:88 for issue #2167's auto-sync-scope correction - none of the lines here overlap with that diff. Refs #2183, #2167 Signed-off-by: Martin Vogel <martin.vogel.tech@gmail.com>
DeusData
force-pushed
the
fix/issue-2167
branch
from
September 30, 2026 22:34
6469d3b to
e5bd709
Compare
This branch has not been deployed
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.
The background watcher only watches the project an open MCP session is
rooted in, and drops the watch when the last session for that project
closes. A repository indexed with
cli index_repository(or from asession rooted elsewhere) is never watched, but README promised
"Auto-sync keeps it fresh after that" for every index_repository target
and the MCP instructions said "Indexes auto-refresh". Nothing in the
status output said which projects were actually watched, so a stale
CLI-indexed repo looked like a watcher bug.
What gets watched is unchanged. Instead:
watchobject:watched, plusstrategy,poll_interval_msandlast_scan_atwhen watched, or areason(not_session_project, cli_session, auto_watch_off, watcher_disabled,
not_registered, no_watcher) and a re-index hint when not. It does not
touch the feat(mcp): report truthful index freshness from checkout evidence #1561
freshnesskey.scan time through atomics and exposes them via
cbm_watcher_project_info(); the daemon answers index_status through a
session-scoped provider.
description and the MCP instructions now state the real scope.
watcher_enabledalready shows inconfig listsince v0.11.0 (680bc72).Adds a
watchobject to index_status next to #2346'scross_repoobject (separate hunks). Follow-up noted, not in this PR: README.md:132 still saysclinever connects to the daemon, which is stale on main.Fixes #2167