GitHub auto-deploys drop service rootDirectory and railway.json after 2026-08-17 incident
aurevan-vourthen
PROOP

2 months ago

GitHub auto-deploys for a monorepo service stopped applying stored rootDirectory and railway.json during the 2026-08-17 GitHub/Railway incidents and have not recovered.

Service settings: rootDirectory=/backend, railwayConfigFile=/backend/railway.json, builder=NIXPACKS. Auto-deploy is on. Railway GitHub App has code access, access in the selected repo, no pending permissions banner. Reconnected the repo in Source. Same break on sibling service (was DOCKERFILE at root file, now RAILPACK with no root) at the same timestamps.

Last good GitHub deploy: 2026-08-17T02:52:21Z, NIXPACKS, root=/backend, configFile=/backend/railway.json.

First bad GitHub deploy: 2026-08-17T15:24:08Z, RAILPACK, root empty, configFile empty. That is during GitHub incident zkxwbgr0cnmx (archive/raw downloads ~50% error) and Railway GitHub Auto-Deploys partial outage.

Later GitHub / latest-commit deploys, all missing root and configFile. CLI railway up from repo root still works. CLI user gets Not Authorized on gitHubRepoAccessAvailable / githubRepo / githubWritableScopes. environmentTriggersDeploy previously returned “Resource not accessible by integration.”

Need GitHub sourced deploys to apply the stored root and config file again, as they did before 15:24 UTC on Aug 17.

Solved$20 Bounty

0 Replies

Status changed to Awaiting Railway Response Railway • about 2 months ago


Railway
BOT

2 months ago

This thread has been opened as a bounty so the community can help solve it.

Status changed to Open Railway • about 2 months ago


Status changed to Solved aurevan-vourthen • about 2 months ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...