{"version":"2.1.300","anchor":"self-hosted-runner-orchestrator-log-writer-and-fatal-exit-re","canonical_anchor":"self-hosted-runner-orchestrator-log-writer-and-fatal-exit-re","heading":"Self-hosted runner logs now go through one shared log writer","tier":"internal","area":"Self-Hosted Runner","scope":"org","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300\/e\/self-hosted-runner-orchestrator-log-writer-and-fatal-exit-re","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300","markdown":"### Self-hosted runner logs now go through one shared log writer\n\nThe self-hosted runner's orchestrator now sends its logs through one shared writer, described in a `[runner:log]` debug line at startup\n\n**Unclear.** Where the shared log writer sends logs is not shown.\n\n**What**\n\nWhen you run Claude Code on your own runner machines, a coordinating process called the orchestrator manages the work. It now sends its log messages through one shared log writer instead of printing them directly. At startup it writes a debug line starting with `[runner:log]` that describes this writer. The steps that stop the runner after a fatal error, and that handle environment secrets and the base folder, also log through this writer now.\n\n**Why**\n\nWhere the orchestrator's logs end up may be different, so check the `[runner:log]` line at startup to see how logging is set up.\n\n- Area: Self-Hosted Runner\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: org\n- Heads-up: no"}