ETIMEDOUT: request timed out
A socket exceeded its timeout before the response arrived.
28 issues · 9 open · 19 resolved (68%) · first seen Feb 26, 2025
First response · network
What to do now
Most likely: A DNS, proxy, TLS or socket layer interrupted the connection.
- Check whether the same request works without the current VPN or corporate proxy.
- Verify DNS and certificate trust without disabling TLS verification.
- Capture the hostname and exact socket code before escalating.
These are conservative triage steps, not an official Anthropic fix. Use the issue and workaround evidence below before making a destructive configuration change.
Is this getting better or worse?
This class of problem is holding steady. 11 new reports in the last 90 days vs 11 in the 90 before — 0%. The open backlog peaked at 9 in 2026-08 and sits at 9 today.
Derived by counting each matching issue from its creation date until it closed. First-party
computation over the anthropics/claude-code tracker — the trend is not published
anywhere upstream.
Which builds report this
18 of these reports name the Claude Code build they were running, spanning 16 releases. Heaviest reporters:
- Claude Code v2.1.234
- Claude Code v2.1.81
- Claude Code v2.1.232
- Claude Code v2.1.224
- Claude Code v2.1.220
- Claude Code v2.1.218
- Claude Code v2.1.209
- Claude Code v2.1.187
Report volume, not a defect count: an issue can mention a build without that build having caused the failure. Versions come from the version string in the bug-report template, matched against real published tags.
Do these get fixed?
Median time to close is 29 days across 19 closures. Of the 19 closures with a recorded reason, 11% were closed as completed and 17 as not-planned or duplicate. The most recent completed fix landed in 2026-03.
Workarounds reported by the community
6 of these issues post a workaround someone says worked. The highest-engagement ones:
- [BUG] WSL: "Unable to connect to Anthropic services" resolved by disabling telemetry
- …pected) - and both work - SSL/TLS handshake works: succeeds Workaround that works: `bash DISABLE TELEMETRY=1 claude
Found in the description of #2481 · resolved - Network Timeout Errors Occurring Frequently in API Calls
- …do. Sometimes, restarting the binary helps. Update: Another workaround is to open multiple windows—4 work for me. Run in each. When one starts working well, close the others.
Found in the thread of #128 · resolved - Telegram plugin: polling exits permanently on network errors during long sessions
- …untu 25.04 - Session duration when issue occurs: 4-8+ hours Workaround Restarting the Claude Code session re-establishes the polling connection. Using with helps for daemon-mode agents.
Found in the description of #40603 · resolved - [BUG] Claude Code cannot connect - Cloudflare TLS fingerprint rejection (ECONNRESET/ETIMEDOUT)
- …oesn't expose TLS configuration via NODE OPTIONS. Attempted Workarounds (all failed for Claude Code) - — no effect - — no effect (undici ignores it) - — no effect - — not tested (security concern) - — updated to v2.1.81,…
Found in the description of #37300 · resolved - [BUG] Background workers (bg-spare) never release memory after jobs complete — monotonic RSS growth saturates 24GB RAM in ~7 days on always-on machines
- …urns after jobs complete; extrapolates to full RAM in days. Workaround we use When all of a project's jobs are : SIGTERM its claimed + parent pairs. Job records, transcripts (jsonl), and the FleetView dashboard are unaff…
Found in the description of #88116 · still open
Excerpts are quoted from the linked issue so you can judge relevance before opening it. Threads that only ask for a workaround, or report that there is none, are excluded. These are community suggestions, not official guidance from Anthropic — read the full thread before applying one.
Matching issues 1–28
Ranked by community engagement (reactions weighted over comments).
- [Bug] Claude is being lazy - very very lazy
- [BUG] WSL: "Unable to connect to Anthropic services" resolved by disabling telemetry
- Network Timeout Errors Occurring Frequently in API Calls
- [BUG] Connection-level errors (ECONNRESET, EPIPE, ETIMEDOUT) are never retried — immediate failure with zero recovery
- I'm ready to help generate GitHub issue titles for Claude Code bug reports. Please share the bug report details, and I'll create a concise, technical title following the format specified.
- [Feature] OnApiRetry hook event for observability and control over internal retry behavior
- I appreciate the feedback, but I need the actual bug report or issue details to generate a GitHub issue title. Could you please provide: 1. A description of the problem you encountered 2. Any error messages or unexpected behavior 3. Steps to reproduce the
- [BUG] Native binary hangs indefinitely with 100% CPU futex spin loop on startup (VPS/virtualized environment)
- [Bug] Scroll position jumps to top on streamed output (Linux regression)
- Bug: ECONNRESET ('terminated') during streaming with partial data kills the session — no retry, no fallback
- [BUG] Hung tool calls should be cancellable — a timeout or manual cancel button per tool call
- [Bug] AI Agent Executing Destructive Operations on Production Environment Without Safeguards
- [Bug] Claude model degradation: Memory and consistency issues across files
- Telegram plugin: polling exits permanently on network errors during long sessions
- ECONNRESET kills session — needs auto-retry/reconnect
- [Bug] Anthropic API Error: Missing thinking block in assistant message with extended thinking enabled
- i recently travelled to the uae from uganda , and am...
- Background Bash tasks get lost/duplicated, turning a simple deploy into 10+ hours of retries
- [BUG] Claude Code cannot connect - Cloudflare TLS fingerprint rejection (ECONNRESET/ETIMEDOUT)
- [BUG] Background workers (bg-spare) never release memory after jobs complete — monotonic RSS growth saturates 24GB RAM in ~7 days on always-on machines
- [BUG] Claude Code CLI native binary spins at ~100% CPU indefinitely on startup (VS Code Remote-SSH, Ubuntu 26.04)
- [BUG] macOS DarkWake: transient DNS failure during OAuth refresh loses login
- [DOCS] Connection-drops section omits that partial responses are preserved and the spinner no longer gets stuck at "running tool" (v2.1.179)
- [Bug] Anthropic API Error: HTTP 500 and 529 timeout during sessions
- CLI hangs indefinitely on every subcommand except --version (headless Linux, both native and npm installs)
- Telegram channel plugin: poller gives up on 409 after ~28 s and never recovers, while /mcp still reports "connected"
- [BUG] Error: Claude Code process terminated by signal SIGKILL
- Auto-reconnect after wake exhausts after 3 attempts (~6s), no further retry