Skip to content

Commit 45fdebf

Browse files
committed
feat: Enhance task ID overrides for pod name and validation tasks in tpu-info DAG
1 parent 2807c4d commit 45fdebf

1 file changed

Lines changed: 14 additions & 8 deletions

File tree

dags/tpu_observability/tpu_info_cli_validation_dags.py

Lines changed: 14 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -169,11 +169,11 @@ def validate_process(info, pod_name: str) -> str:
169169
tpu_topology=config.tpu_topology,
170170
)
171171

172-
create_node_pool = node_pool.create(
172+
create_node_pool = node_pool.create.override(task_id="create_node_pool")(
173173
node_pool=cluster_info,
174174
)
175175

176-
apply_time = jobset.run_workload(
176+
apply_time = jobset.run_workload.override(task_id="run_workload")(
177177
node_pool=cluster_info,
178178
yaml_config=jobset_config.generate_yaml(
179179
workload_script=Workload.JAX_TPU_BENCHMARK
@@ -195,16 +195,22 @@ def validate_process(info, pod_name: str) -> str:
195195
with TaskGroup( # pylint: disable=unexpected-keyword-arg
196196
group_id="verification_group"
197197
) as verification_group:
198-
help_validation = validate_help.partial(info=cluster_info).expand(
199-
pod_name=pod_names
198+
help_validation = (
199+
validate_help.override(task_id="validate_help")
200+
.partial(info=cluster_info)
201+
.expand(pod_name=pod_names)
200202
)
201203

202-
version_validation = validate_version.partial(info=cluster_info).expand(
203-
pod_name=pod_names
204+
version_validation = (
205+
validate_version.override(task_id="validate_version")
206+
.partial(info=cluster_info)
207+
.expand(pod_name=pod_names)
204208
)
205209

206-
process_validation = validate_process.partial(info=cluster_info).expand(
207-
pod_name=pod_names
210+
process_validation = (
211+
validate_process.override(task_id="validate_process")
212+
.partial(info=cluster_info)
213+
.expand(pod_name=pod_names)
208214
)
209215

210216
cleanup_workload = jobset.end_workload.override(

0 commit comments

Comments
 (0)