[cfremling]

seq stop reached camerad only while the EXPOSE wait was set, i.e. during sub-exposure 1 of an NEXP>1 series, and after an abort camerad could still start the next sub-exposure.

• sequencerd accepts stop for the whole series of the current target (after the first sub-exposure, camerad not READY covers the rest) and no longer clears a cancel.
• camerad ends the series after the frame in progress, cut short if it is integrating with 5 s or more left, instead of starting the next sub-exposure. An abort sets the remaining count to zero, and a continuation cannot undo a stop.
• A cancelled exposure's remaining PIXELCOUNT messages cannot set the READOUT wait later.
• A command that finds a frame write still finishing (READY is published a few ms before the last write ends) waits up to 2 s for it instead of being refused.

Stacked on #500 (base branch feat/auto-ontarget-ring); uses #499's readout state.

Jira: PAL-142. Not yet tested on sky with NEXP>1: all 18 exposures on 2026-10-28/29 were NEXP=1. A stop on a single exposure that night ended the integration as expected.

🤖 Generated with Claude Code

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	🔄 Running since 2026-10-29T14:15:34.781751Z	7a22f56	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[cfremling]

Combined into #503 with the changes it depends on, so that every open PR applies to main on its own; its change is one commit of #503. Closing.