#829 Fix infinite loop in log tail retry logic Pull request opened by pingsutw Summary • Only retry on

NOT_FOUND

gRPC errors when tailing logs in

Logs.tail()

• Non-

NOT_FOUND

errors (e.g.

UNAVAILABLE

,

INTERNAL

) are now re-raised immediately instead of silently looping forever • Previously, after exhausting retries on non-

NOT_FOUND

errors, the code fell through both `if`/`else` branches and looped infinitely in

while True

without any sleep, causing

flyte get logs

to hang Test plan • Run

flyte get logs <run> <action>

against an action with available logs — should stream normally • Run

flyte get logs

against a non-existent action — should raise

LogsNotYetAvailableError

after retries • Simulate a non-NOT_FOUND gRPC error — should raise immediately instead of hanging flyteorg/flyte-sdk