[BUG] Upstream crashes On May15th-16th
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
Diagnostics
└ Currently running: package-manager (2.1.56)
└ Package manager: homebrew
└ Path: /opt/homebrew/Caskroom/claude-code/2.1.56/claude
└ Invoked: /opt/homebrew/Caskroom/claude-code/2.1.56/claude
└ Config install method: global
└ Search: OK (bundled)
Updates
└ Auto-updates: Managed by package manager
└ Auto-update channel: latest
└ Stable version: 2.1.133
└ Latest version: 2.1.143
"Repeated sdk_stream_ended_no_result with had_error: true on Opus 4.7, May 15–16."
A few x-client-request-id values (one example: 69bbe7d6-b2c3-45c8-b25a-0c43d5c8f0e0 at 2026-05-15T23:00:47Z).
The stall pattern: stream opens, 600–700 bytes through, then idle ≥15s, then had_error.
Effective window observed: 980000 (1M-context path).
The /doctor output pasted in.
The financial impact: usage limits exhausted from billed-but-failed streams. State plainly that you're requesting credit.
What Should Happen?
I'm expecting to continue work uninterrupted
Error Messages/Logs
Steps to Reproduce
Regular dotnet development during a session
Claude Model
Opus
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
2.1.56
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
_No response_
This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗