Skip to content

SPEC 7.6: the reconnect Gap is closed by plugrl-env-client#15; a dated correction in its place #12

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tactino merged 1 commit into from
Oct 2, 2026
Merged

SPEC 7.6: the reconnect Gap is closed by plugrl-env-client#15; a dated correction in its place #12

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 11 additions & 12 deletions SPEC.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -608,18 +608,17 @@ environment it has no step state for. That condition has exactly one cause -
the connection was replaced mid-run - and storing the transition silently
puts a hole in the training data that nothing downstream can detect.

> **Gap - the reference client breaks this on one path.**
> `plugrl-env-client` run with `--reconnect-on-server-stop` resends the
> `feedback` it was holding instead of dropping it, when the close it hit was
> the server's `plugrl-server-stop`. In `websocket_env_client_agent.py`,
> `feedback()`'s `SERVER_STOP_REASON` branch `continue`s its retry loop,
> which opens a new connection, reads fresh metadata, and re-sends the same
> payload - the exact resend the historical note below says this section
> exists to prevent. Every other close path in that method returns and drops
> the transition, and `tests/test_reconnect_drops_feedback.py` covers three
> of them (a keepalive timeout, a plain 1000 close, a resync) but not this
> one. It is the only path in the reference client that violates the **MUST**
> above.
> **Correction, 2026-10-01.** This section carried a Gap until today:
> `plugrl-env-client` run with `--reconnect-on-server-stop` resent the
> `feedback` it was holding after a `plugrl-server-stop` close, instead of
> dropping it. plugrl-env-client#15 removed that path. `feedback()` never
> reconnects now: it raises `ConnectionReplaced`, and the rollout drops every
> chunk in flight and asks again for all its environments. The Gap was also
> incomplete. The probe checker (section 8.1) found a second violation it did
> not mention: after any reconnect, environments in the middle of a chunk
> kept executing it, then sent its `feedback` on the new connection. #15
> fixed that too. `plugrl-conformance --probe --scenario resync` checks both,
> and plugrl-env-client's CI runs it.

> **Where reconnects come from.** Nothing in this protocol causes them and
> nothing in it can prevent them: a suspended laptop, a flaky link, an
Expand Down
Loading