sunank200 commented on code in PR #39130:
URL: https://github.com/apache/airflow/pull/39130#discussion_r1575661892
##########
airflow/providers/google/cloud/triggers/dataproc.py:
##########
@@ -140,24 +142,97 @@ def serialize(self) -> tuple[str, dict[str, Any]]:
"gcp_conn_id": self.gcp_conn_id,
"impersonation_chain": self.impersonation_chain,
"polling_interval_seconds": self.polling_interval_seconds,
+ "delete_on_error": self.delete_on_error,
},
)
async def run(self) -> AsyncIterator[TriggerEvent]:
- while True:
- cluster = await self.get_async_hook().get_cluster(
- project_id=self.project_id, region=self.region,
cluster_name=self.cluster_name
+ """Run the trigger."""
+ try:
+ while True:
+ cluster = await self.fetch_cluster_status()
+ if self.check_cluster_state(cluster.status.state):
+ if cluster.status.state == ClusterStatus.State.ERROR:
+ await self.gather_diagnostics_and_maybe_delete(cluster)
+ else:
+ yield TriggerEvent(
+ {
+ "cluster_name": self.cluster_name,
+ "cluster_state": cluster.status.state,
+ "cluster": cluster,
+ }
+ )
+ break
+ self.log.info("Sleeping for %s seconds.",
self.polling_interval_seconds)
+ await asyncio.sleep(self.polling_interval_seconds)
+ except asyncio.CancelledError:
+ await self.handle_cancellation()
+
+ async def fetch_cluster_status(self) -> Cluster:
+ """Fetch the cluster status."""
+ return await self.get_async_hook().get_cluster(
+ project_id=self.project_id, region=self.region,
cluster_name=self.cluster_name
+ )
+
+ def check_cluster_state(self, state: ClusterStatus.State) -> bool:
Review Comment:
It checks two cases - if it is running or it is in error state.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
