Codex App window repeatedly turns completely blank after renderer crashes
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.5k
- PR merge metrics
- PR metrics pending
Description
What version of the Codex App are you using (From “About Codex” dialog)?
26.803.61601 (build 6396); bundled codex-cli 0.147.0-alpha.6.5
What subscription do you have?
ChatGPT subscription (exact tier is not displayed in the Codex App)
What platform is your computer?
macOS 26.5.1 (Build 25F80), Apple Silicon arm64; Darwin 25.5.0
What issue are you seeing?
The Codex App window intermittently becomes completely blank while multiple important tasks are running. Restarting the app restores the UI, but the interruption has caused important tasks to be left unfinished or difficult to recover.
Local Crashpad records show repeated renderer crashes. On 2026-08-11 (Asia/Shanghai), renderer crash records were created at 16:09:11, 16:12:59, and 16:17:31. The same machine also has renderer crash records from 2026-08-10 at 10:42:54 and 20:24:25.
This appears to be a UI renderer failure rather than a project/workspace error. No project names, task content, account identifiers, or raw logs are included here.
What steps can reproduce the bug?
The issue is intermittent, but the most consistent sequence observed in the local app logs is:
- Run multiple Codex tasks concurrently.
- At least one task starts or uses the Codex in-app Browser.
- Within several seconds, a renderer crash is recorded.
- The app UI may become entirely blank.
- Quit and restart the Codex App; the UI returns.
Examples on 2026-08-11 (Asia/Shanghai):
- 16:09:04 in-app Browser session startup → 16:09:11 renderer crash.
- 16:12:44 in-app Browser session startup → 16:12:59 renderer crash.
- 16:17:28 in-app Browser session startup → 16:17:31 renderer crash.
After these events, app logs report messages such as:
route window is not liveNo ChatGPT browser route is availablehost did not respond to sync request
No system memory-pressure or jetsam event was recorded at the incident time. Restarting the app is the only reliable recovery I have found.
What is the expected behavior?
A renderer or in-app Browser failure should not blank the entire Codex App. If a child renderer crashes, the app should automatically recover the affected view while keeping task state and active runs accessible.
Additional information
Impact: High. This has happened repeatedly and has interrupted important long-running work, leaving some tasks unfinished.
The app updated/relaunched around 16:11 on 2026-08-11, but renderer crashes occurred both before and after that update, so the update alone does not explain the issue.
The app was using substantial memory at the time, but the macOS system had no recorded memory-pressure event and still had ample available memory. The strongest correlation is with in-app Browser session startup.
I can provide the relevant redacted app-log excerpts privately if maintainers need them.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the Codex App renderer and in-app Browser flow, using the reported Crashpad records and app-log messages such as route window is not live. Reproduce with multiple concurrent tasks and Browser startup on macOS, then trace the renderer-crash path. Done means a child renderer or Browser failure no longer blanks the whole window and active tasks remain accessible without restarting.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- macos
- Domain
- desktop
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100