Claude Desktop conversations stopping due to internal server errors

[somethingwithproof]

Description

Claude Desktop conversations are stopping unexpectedly with internal server errors from the backend. The errors are marked as non-retryable, causing immediate conversation termination and message store sync loss.

Environment

• Platform: macOS Darwin 25.2.0
• Date: 2026-01-23
• Claude Desktop Version: 2.1.8 (from logs)

Log Evidence

From ~/Library/Logs/Claude/claude.ai-web.log:

2026-01-23 00:03:34 [error] [COMPLETION] Request failed Error: Internal server error
2026-01-23 00:03:34 [error] [COMPLETION] Not retryable error, throwing
2026-01-23 00:03:34 [error] Uncaught (in promise) Error: Internal server error
2026-01-23 00:03:34 [warn] [COMPLETION] message_store_sync_loss [object Object]

2026-01-23 00:04:00 [error] [COMPLETION] Request failed Error: Internal server error
2026-01-23 00:04:00 [error] [COMPLETION] Not retryable error, throwing
2026-01-23 00:04:00 [error] Uncaught (in promise) Error: Internal server error
2026-01-23 00:04:00 [warn] [COMPLETION] message_store_sync_loss [object Object]

2026-01-23 00:04:27 [error] [COMPLETION] Request failed Error: Internal server error
2026-01-23 00:04:27 [error] [COMPLETION] Not retryable error, throwing
2026-01-23 00:04:27 [error] Uncaught (in promise) Error: Internal server error
2026-01-23 00:04:27 [warn] [COMPLETION] message_store_sync_loss [object Object]

2026-01-23 00:14:09 [error] [COMPLETION] Request failed Error: Internal server error
2026-01-23 00:14:09 [error] [COMPLETION] Not retryable error, throwing
2026-01-23 00:14:15 [error] Uncaught (in promise) Error: Internal server error
2026-01-23 00:14:15 [warn] [COMPLETION] message_store_sync_loss [object Object]

2026-01-23 00:19:52 [error] [COMPLETION] Request failed Error: Internal server error
2026-01-23 00:19:52 [error] [COMPLETION] Not retryable error, throwing
2026-01-23 00:19:52 [error] Uncaught (in promise) Error: Internal server error
2026-01-23 00:19:52 [warn] [COMPLETION] message_store_sync_loss [object Object]

Pattern

Multiple internal server errors occurring within short time windows (clustered around 00:03-00:04 and 00:14-00:19), all resulting in:

1. Non-retryable error classification
2. Immediate conversation termination
3. Message store sync loss

Impact

• Conversations stop abruptly without user-visible error messages
• Work in progress is lost
• User must restart conversation manually

Expected Behavior

• Transient server errors should be retried
• User should receive clear error message when conversation must stop
• Message store should maintain sync or gracefully degrade

Suggested Fix

1. Make internal server errors retryable with exponential backoff
2. Show user-facing error message explaining the issue
3. Preserve conversation state even when sync is lost

[github-actions]

Found 3 possible duplicate issues:

1. Unrecoverable 500 errors requiring /clear or new session #19330
2. API 500 Error Loop After Cumulative Context Exceeds Threshold #15076
3. SessionEnd hooks don't fire on API 500 errors #20197

This issue will be automatically closed as a duplicate in 3 days.

• If your issue is a duplicate, please close it and 👍 the existing issue instead
• To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

[github-actions]

This issue has been automatically closed as a duplicate of #19330.

If this is incorrect, please re-open this issue or create a new one.

🤖 Generated with Claude Code

[github-actions]

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.