claude-code - 💡(How to fix) Fix [Bug] Subagent dispatch fails with "Extra usage required for 1M context" on Max plan despite included feature [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#54948Fetched 2026-05-01 05:50:16
View on GitHub
Comments
1
Participants
2
Timeline
7
Reactions
0
Author
Timeline (top)
labeled ×5commented ×1cross-referenced ×1

Error Message

[{"error":"Error: NON-FATAL: Lock acquisition failed for /Users/goos/.local/share/claude/versions/2.1.123 (expected in multi-process scenarios)\n at x86 (/$bunfs/root/src/entrypoints/cli.js:2769:2177)\n at Jw8 (/$bunfs/root/src/entrypoints/cli.js:2769:1257)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T08:56:33.027Z"},{"error":"Error: Request was aborted.\n at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:3448)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T08:56:41.297Z"}]

Root Cause

Cross-references Already-filed issues with the same root cause: #44117, #45390, #46780, #46967. #44117 is the closest match. All still affect 2.1.123. Please consider this a +1 with the additional data point that 2.1.123 does not fix the regression.

Fix Action

Fix / Workaround

Symptom Every Task-tool subagent dispatch (in our case manager-tdd and expert-backend) returns within ~1s with: API Error: Extra usage is required for 1M context · run /extra-usage to enable, or /model to switch to standard context The same error blocks WebSearch from subagent contexts. The parent session itself is fine — only delegated agents fail.

Actual

  • 100% failure rate on subagent dispatch.
  • Failure happens before the agent's first tool call — the prompt is rejected at request time.
  • Parent session continues to work, including direct WebSearch from parent.

Repro

  1. Max plan, Opus 4.7 selected via /model. 2. Ensure /extra-usage is OFF.
  2. From a fresh session, dispatch any subagent. 4. Observe immediate error.

Code Example

[{"error":"Error: NON-FATAL: Lock acquisition failed for /Users/goos/.local/share/claude/versions/2.1.123 (expected in multi-process scenarios)\n    at x86 (/$bunfs/root/src/entrypoints/cli.js:2769:2177)\n    at Jw8 (/$bunfs/root/src/entrypoints/cli.js:2769:1257)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T08:56:33.027Z"},{"error":"Error: Request was aborted.\n    at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:3448)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T08:56:41.297Z"}]
RAW_BUFFERClick to expand / collapse

Bug Description [Subject] Subagent + WebSearch fail on Max plan with "Extra usage is required for 1M context" — regression

Environment

  • Claude Code: 2.1.123 (macOS, Apple Silicon)
  • Plan: Max ($200/mo)
  • Model: claude-opus-4-7 (Opus 4.7, 1M context)
  • Effort level: xhigh
  • Extra-usage toggle: NOT enabled (intentionally — 1M is supposed to be included on Max)

Symptom Every Task-tool subagent dispatch (in our case manager-tdd and expert-backend) returns within ~1s with: API Error: Extra usage is required for 1M context · run /extra-usage to enable, or /model to switch to standard context The same error blocks WebSearch from subagent contexts. The parent session itself is fine — only delegated agents fail.

Expected Per the March 13, 2026 announcement and the in-app docs, 1M context on Opus 4.6 / 4.7 is INCLUDED in Max plan with no extra-usage toggle required. Subagents should inherit the parent's plan coverage, not require a separate per-call extra-usage check.

Actual

  • 100% failure rate on subagent dispatch.
  • Failure happens before the agent's first tool call — the prompt is rejected at request time.
  • Parent session continues to work, including direct WebSearch from parent.

Repro

  1. Max plan, Opus 4.7 selected via /model. 2. Ensure /extra-usage is OFF.
  2. From a fresh session, dispatch any subagent. 4. Observe immediate error.

Cross-references Already-filed issues with the same root cause: #44117, #45390, #46780, #46967. #44117 is the closest match. All still affect 2.1.123. Please consider this a +1 with the additional data point that 2.1.123 does not fix the regression.

Impact SPEC-driven workflows that rely on subagent delegation are completely blocked on Max plan despite 1M being a paid feature. Users either downgrade context or pay extra-usage on top of Max (double-billing).

Request Treat as a regression. Propagate the parent session's "1M is included on Max" flag to spawned subagent requests so they don't trip the per-call extra-usage gate.

Environment Info

  • Platform: darwin
  • Terminal: tmux
  • Version: 2.1.123
  • Feedback ID: fa9b09df-f5f8-4062-ac70-f4481cbc65ad

Errors

[{"error":"Error: NON-FATAL: Lock acquisition failed for /Users/goos/.local/share/claude/versions/2.1.123 (expected in multi-process scenarios)\n    at x86 (/$bunfs/root/src/entrypoints/cli.js:2769:2177)\n    at Jw8 (/$bunfs/root/src/entrypoints/cli.js:2769:1257)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T08:56:33.027Z"},{"error":"Error: Request was aborted.\n    at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:3448)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T08:56:41.297Z"}]

extent analysis

TL;DR

The issue can be resolved by propagating the parent session's "1M is included on Max" flag to spawned subagent requests.

Guidance

  • The error message "Extra usage is required for 1M context" suggests that the subagent requests are being treated as separate entities, rather than inheriting the parent session's plan coverage.
  • To verify this, try enabling the extra-usage toggle for the subagent requests and see if the error persists.
  • The fact that the parent session continues to work, including direct WebSearch, implies that the issue is specific to subagent delegation.
  • Reviewing the cross-referenced issues (#44117, #45390, #46780, #46967) may provide additional insight into the root cause of the problem.

Example

No code snippet is provided as the issue is more related to the configuration and plan coverage.

Notes

The issue seems to be a regression, and the provided version (2.1.123) does not fix the problem. The error messages in the Errors section do not seem to be directly related to the main issue.

Recommendation

Apply a workaround by treating subagent requests as part of the parent session's plan coverage, until a fixed version is released. This is because the current version (2.1.123) does not resolve the issue, and downgrading or paying extra-usage on top of Max plan is not a viable solution.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

claude-code - 💡(How to fix) Fix [Bug] Subagent dispatch fails with "Extra usage required for 1M context" on Max plan despite included feature [1 comments, 2 participants]