Skip to content

[P2] Investigate idle disconnects during long-running exec streams #20

Open

Description

@brendandburns

Upstream: kubernetes-client#1841
Priority: P2 (stream reliability; current reproduction required)

Long-running exec commands are reported to lose their WebSocket connection after approximately 300 seconds without output, even though the remote process remains running. The report suggests missing keepalive pings, but proxy/server idle limits must be distinguished from a client defect.

Reproduce with a controlled idle timeout and evaluate existing ping/pong, connection-health, and timeout behavior. If a keepalive mechanism is needed, follow existing configuration patterns and avoid introducing unbounded background activity.

Acceptance criteria:

  • Cover keepalive/connection lifecycle behavior with deterministic tests rather than multi-minute sleeps.
  • Preserve deliberate timeout, cancellation, and close behavior.
  • Explain client versus proxy/server limitations and actionable failure reporting.

Activity

  1. linked a pull request that will close this issueKeep idle exec WebSocket streams alive #35on Nov 19, 2026
  2. brendandburns commented on Nov 19, 2026

    @brendandburns
    OwnerAuthor

    Clarification to the original assignment: this task's designated agent/base branch is agent/p2-issue-20. Raise/target the PR against that branch, not master; the existing draft PR has been retargeted accordingly. Keep the requested commit order: first a test-only commit demonstrating the issue, then a separate fix commit only if reproduced. If it does not reproduce, leave the evidence/test commit and explain why no fix is appropriate.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

bugSomething isn't working

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions