claude-code - 💡(How to fix) Fix [BUG] [Windows] 1.2773.0 browser-process V8 OOM on launch [1 comments, 2 participants]

Crashpad dump signature:

• process_type = browser
• Electron 41.2.0
• electron.v8-oom.location = CALL_AND_RETRY_LAST
• electron.v8-oom.is_heap_oom
• OOM error in V8: CALL_AND_RETRY_LAST Allocation failed - JavaScript heap out of memory

main.log also shows:

• UnknownError: Internal error opening backing store for indexedDB.open

cowork-service.log still shows the app connects, signature-verifies, then drops with EOF shortly after. The VM service stays healthy.

I also tested with:

• VM bundle compression fixed/uncompressed
• transcript/session caches moved aside
• Chromium caches cleared
• broad Claude profile reset, leaving only claude_desktop_config.json and config.json
• CoworkVMService running normally

The crash still happens from that near-empty state.

TL;DR

The issue is likely related to a memory leak or excessive memory usage in the Electron application, causing it to crash on launch.

Guidance

• Investigate the electron.v8-oom.location = CALL_AND_RETRY_LAST error in the Crashpad dump, which suggests a JavaScript heap out of memory issue.
• Check the main.log for any other errors or warnings that may indicate the cause of the UnknownError: Internal error opening backing store for indexedDB.open error.
• Consider increasing the memory allocation for the Electron application or optimizing the code to reduce memory usage.
• Verify that the issue is not related to the cowork-service.log or VM service by checking their logs for any errors or anomalies.

Example

No code snippet is provided as the issue is more related to configuration and memory management.

Notes

The issue seems to be specific to the Windows platform and the MSIX install. The fact that it worked in a previous version suggests a regression, but without knowing the last working version, it's difficult to pinpoint the exact cause.

Recommendation

Apply workaround: Increase the memory allocation for the Electron application or optimize the code to reduce memory usage, as the root cause seems to be related to memory management.