fix(windows): stop GUI update from stacking visible terminals#167
Closed
Wibias wants to merge 7 commits into
Closed
fix(windows): stop GUI update from stacking visible terminals#167Wibias wants to merge 7 commits into
Wibias wants to merge 7 commits into
Conversation
…le flashes Pin post-update proxy restart to config.port, wait briefly for the listen socket to free, and avoid spawning visible PowerShell/icacls windows on the start hot path.
Port stickiness stays scoped to restart/port behavior; icacls ACL soft-fail lives elsewhere.
Hide the Task Scheduler wrapper via a WScript launcher (window style 0), pipe npm/update child stdio under OCX_SERVICE=1, and keep post-update restart pinned to config.port so failed starts do not fuel the :loop.
Keep main's absorbed port-stickiness + pid identity checks; retain hidden WScript service launcher and quiet GUI-worker npm stdio.
WaitOnReturn=True so wscript stays alive for the batch lifetime; False made CI see Ready despite a healthy proxy.
lidge-jun
added a commit
that referenced
this pull request
Jul 20, 2026
…child stdio (port #167) - ocx update --help printed nothing useful and ran the REAL self-update, stopping the proxy mid-stream. Both entry paths (Bun CLI and the npm bin launcher, which intercepts update before Bun starts) now print usage and exit 0. Verified live: both print usage, exit 0, no update side effects. - Port the still-unique part of PR #167: under the GUI update worker (OCX_SERVICE=1, stdio ignored) the stop/installer/service-reinstall children now use pipe stdio with bounded output relay, so npm.cmd (shell:true) cannot open stacked visible consoles on Windows. Backend- preserving serviceReinstallArgs() is kept; the rest of #167 is already superseded by the dev Windows-service lane.
Owner
|
Thanks for tracking down the console-storm root cause — the diagnosis in this PR (visible Task Scheduler console + wrapper retry loop) matches what we found independently, and it helped confirm the direction. Since this PR opened, the
One part of this PR was still unique and valuable: piping child stdio under the GUI update worker ( So: closed as superseded, with the stdio fix landed. Thanks again! |
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
2 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
opencodex-service.cmdas a visible console, and the wrapper:loopretriedocx startevery ~5s when start failed mid-update (often port-busy).Run …, 0, False) and point the task atwscript.exe.ocx start --port+ prefer-retry / wait-for-free) from closed fix: keep configured port after update restart #152 so restarts do not hop ports and fuel the loop.OCX_SERVICE=1), use pipe stdio for npm/stop/service children sonpm.cmddoes not open stacked cmd windows.Test plan
bun test tests/service.test.ts tests/update-job.test.ts tests/update-stop-first.test.ts tests/ports.test.tsocx service installis in use: reinstall/update rewrites the hidden VBS+XML; only one wrapper runs