mirror of
https://github.com/neondatabase/neon.git
synced 2026-01-13 16:32:56 +00:00
9609f7547e
## Problem There are a couple of log warnings tripping up `test_timeline_archival_chaos` - `[stopping left-over name="timeline_delete" tenant_shard_id=2d526292b67dac0e6425266d7079c253 timeline_id=Some(44ba36bfdee5023672c93778985facd9) kind=TimelineDeletionWorker\n')](https://neon-github-public-dev.s3.amazonaws.com/reports/pr-10672/13161357302/index.html#/testresult/716b997bb1d8a021)` - `ignoring attempt to restart exited flush_loop 503d8f401d8887cfaae873040a6cc193/d5eed0673ba37d8992f7ec411363a7e3\n')` Related: https://github.com/neondatabase/neon/issues/10389 ## Summary of changes - Downgrade the 'ignoring attempt to restart' to info -- there's nothing in the design that forbids this happening, i.e. someone calling maybe_spawn_flush_loop concurrently with shutdown() - Prevent timeline deletion tasks outliving tenants by carrying a gateguard. This logically makes sense because the deletion process does call into Tenant to update manifests.