Skip to content

Commit f09ddb1

Browse files
authored
Change default log wait timeout to 10s (#232)
1 parent 9073a2c commit f09ddb1

File tree

1 file changed

+4
-2
lines changed

1 file changed

+4
-2
lines changed

nemo_run/run/torchx_backend/launcher.py

Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -122,7 +122,7 @@ def wait_and_exit(
122122
log: bool,
123123
runner: Runner | None = None,
124124
timeout: int = 10,
125-
log_join_timeout: int = 600,
125+
log_join_timeout: int = 10,
126126
) -> specs.AppStatus:
127127
if runner is None:
128128
runner = get_runner()
@@ -161,7 +161,9 @@ def wait_and_exit(
161161
logger.info(f"Job {app_id} finished: {status.state}")
162162

163163
if log_thread and log_thread.is_alive():
164-
logger.debug("Waiting for log thread to complete...")
164+
logger.warning(
165+
f"Waiting for {app_id}'s log thread to complete for {log_join_timeout} seconds..."
166+
)
165167
log_thread.join(timeout=log_join_timeout)
166168
if log_thread.is_alive():
167169
logger.warning("Log thread did not complete within timeout, some logs may be missing")

0 commit comments

Comments
 (0)