Hi, Flow A repeatedly (loop) run Flow B. I want ru...
# ask-communityp
Patrick Tan
04/01/2022, 12:00 PMHi, Flow A repeatedly (loop) run Flow B. I want run next instance of Flow B after current Flow B is completed. I ran below and 2 instances of the same flow run at same time, and flow-id is same. see screenshot.
Copy code
with Flow("parent-flow") as flow:
for i in range(2):
flow_a = create_flow_run(flow_name="livelots flow", project_name="LiveLots-ETL")
wait_for_flow_a = wait_for_flow_run(flow_a, raise_final_state=True)
print(i)a
Anna Geller
04/01/2022, 12:41 PM
A couple of issues here:
1. You can't run for loops in Prefect 1.0. This is possible in Prefect 2.0, but in 1.0, within the Flow, you can only construct your DAG, i.e., call your tasks - no for loops allowed. Technically your for-loop works because it iterates over a static list of values, but still, this is not something you should do in 1.0 - mapping would be a better approach. Check the example below
2. The print statement is also not allowed
Copy code
from prefect import Flow, unmapped
from prefect.tasks.prefect import create_flow_run
from prefect.executors import LocalDaskExecutor
with Flow("parent_flow", executor=LocalDaskExecutor()) as parent_flow:
mapped_flow_run_ids = create_flow_run.map(
flow_name=["flow_name_1", "flow_name_2", "flow_name_3"],
project_name=unmapped("your_project_name"),
)Anna Geller
04/01/2022, 12:45 PM
And actually, regarding your issue: it's most likely due to the idempotency key. If you add
idempotency_key=datetime.now().strftime("%m/%d/%Y:%H:%M:%S") Copy code
with Flow("parent-flow") as flow:
for i in range(2):
flow_a = create_flow_run(flow_name="livelots flow", project_name="LiveLots-ETL", idempotency_key=datetime.now().strftime("%m/%d/%Y:%H:%M:%S"))
wait_for_flow_a = wait_for_flow_run(flow_a, raise_final_state=True)p
Patrick Tan
04/01/2022, 1:11 PMI added the code and now child flow only execute once, I expect it to run twice (for i in range(2))
Patrick Tan
04/01/2022, 1:14 PM@Anna Geller, miss your 1st response, let me review first
k
Kevin Kho
04/01/2022, 1:50 PM
Yeah this looks like an idempotency key issue. The docstring of the idempotency key in the task gives a good description of what is happening.
p
Patrick Tan
04/01/2022, 3:34 PM@Kevin Kho, I have a parent flow with a task that return a list of files, then for each file, I want to run a child flow by passing filename as parameter. I am scratching my head since task is not iterable, do you any idea or sample code ?
k
Kevin Kho
04/01/2022, 3:36 PM
You can use
Copy code
create_flow_run.map()a
Anna Geller
04/01/2022, 3:47 PM
this is without subflows but a very similar use case to yours https://discourse.prefect.io/t/how-to-sequentially-loop-several-cleanup-tasks-over-a-set-of-files-to-process-for-etl/370
Anna Geller
04/01/2022, 3:48 PM
Copy code
from prefect import task, Flow, Parameter
from prefect.executors import LocalExecutor
from datetime import timedelta
@task(max_retries=5, retry_delay=timedelta(minutes=10))
def read_data_from_file(file_name: str):
return "some_data"
@task(max_retries=5, retry_delay=timedelta(minutes=10))
def etl_1(x):
# some cleaning
return x
@task(max_retries=5, retry_delay=timedelta(minutes=10))
def etl_2(x):
# some cleaning
return x
@task(max_retries=5, retry_delay=timedelta(minutes=10))
def etl_15(x):
# some cleaning
return x
with Flow("clean all files", executor=LocalExecutor()) as flow:
all_files = Parameter(
"files", default=["file1.csv", "file2.csv", ..., "file100.csv"]
)
data = read_data_from_file.map(all_files)
transformed_data = etl_1.map(data)
transformed_data = etl_2.map(transformed_data)
transformed_data = etl_15.map(transformed_data)etl_1.map(data)create_flow_run.map(args)p
Patrick Tan
04/01/2022, 6:26 PM@Anna Geller. Can you tell me what's wrong with my code? I did create_flow_run.map for all the files, which will be passed as parameters, then create_flow_run. Sorry I am new on this:
with Flow("parent-flow") as flow:
mapped_flow_run_ids = create_flow_run.map(parameters=[{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/grays/test-live-lots-raw.tsv"},
{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/Blackwell/20220325_180155_Blackwell-DS=420.tsv"}])
flow_a = create_flow_run(flow_id=mapped_flow_run_ids,
flow_name="livelots flow",
project_name="LiveLots-ETL",
idempotency_key=datetime.now().strftime("%m/%d/%Y:%H:%M:%S"))
wait_for_flow_a = wait_for_flow_run(flow_a, raise_final_state=True)
flow.run()
k
Kevin Kho
04/01/2022, 6:31 PM
What is the error here? Still only one waits created?
p
Patrick Tan
04/01/2022, 7:53 PMThis is the error:
[2022-04-01 155049-0400] INFO - prefect.FlowRunner | Beginning Flow run for 'parent-flow'
[2022-04-01 155049-0400] INFO - prefect.TaskRunner | Task 'create_flow_run': Starting task run...
[2022-04-01 155049-0400] INFO - prefect.TaskRunner | Task 'create_flow_run': Finished task run for task with final state: 'Mapped'
[2022-04-01 155049-0400] INFO - prefect.TaskRunner | Task 'create_flow_run[0]': Starting task run...
[2022-04-01 155049-0400] ERROR - prefect.TaskRunner | Task 'create_flow_run[0]': Exception encountered during task execution!
Traceback (most recent call last):
File "/Users/patricktan/bitbucket/data/live-lots-etl/venv-livelotsetl/lib/python3.7/site-packages/prefect/engine/task_runner.py", line 880, in get_task_run_state
logger=self.logger,
File "/Users/patricktan/bitbucket/data/live-lots-etl/venv-livelotsetl/lib/python3.7/site-packages/prefect/utilities/executors.py", line 468, in run_task_with_timeout
return task.run(*args, **kwargs) # type: ignore
File "/Users/patricktan/bitbucket/data/live-lots-etl/venv-livelotsetl/lib/python3.7/site-packages/prefect/tasks/prefect/flow_run.py", line 119, in create_flow_run
"Both
flow_idflow_nameflow_idflow_nameflow_idflow_nameflow_idflow_namePatrick Tan
04/01/2022, 8:00 PMWhen running child flow, flow name and flow project are same, only parameters are different. So I only map parameters:
Copy code
mapped_flow_run_ids = create_flow_run.map(parameters=[{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/grays/test-live-lots-raw.tsv"},
{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/Blackwell/20220325_180155_Blackwell-DS=420.tsv"}])k
Kevin Kho
04/01/2022, 8:06 PM
create_flow_runp
Patrick Tan
04/01/2022, 8:27 PMI provides create_flow_run.map with flow_name, flow_project and parameter (parameter is list of 2 dictionaries. Then I follow by create_flow_run with flow_id from create_flow_run.map and idempotency_key. The error shows failure to run child flow twice
Copy code
with Flow("parent-flow") as flow:
#filelist = get_filelist_task()
mapped_flow_run_ids = create_flow_run.map(flow_name="livelots flow",
project_name="LiveLots-ETL",
parameters=[{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/grays/test-live-lots-raw.tsv"},
{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/Blackwell/20220325_180155_Blackwell-DS=420.tsv"}]
)
flow_a = create_flow_run(flow_id=mapped_flow_run_ids,
idempotency_key=datetime.now().strftime("%m/%d/%Y:%H:%M:%S"))
wait_for_flow_a = wait_for_flow_run(flow_a, raise_final_state=True)
flow.run()k
Kevin Kho
04/01/2022, 8:30 PM
This looks off because
Copy code
create_flow_run.map(flow_name="livelots flow",
project_name="LiveLots-ETL", Copy code
create_flow_run.map(flow_name=unmapped("livelots flow"),
project_name=("LiveLots-ETL"),from prefect import unmappedp
Patrick Tan
04/01/2022, 8:47 PMThere is some success with using unmapped function. The child flow ran twice with 2 set of parameters (expected). However, they are run at same time with same flow-id, I expected running one at a time. and also get some error message
Copy code
with Flow("parent-flow") as flow:
#filelist = get_filelist_task()
mapped_flow_run_ids = create_flow_run.map(flow_name=unmapped("livelots flow"),
project_name=unmapped("LiveLots-ETL"),
idempotency_key=datetime.now().strftime("%m/%d/%Y:%H:%M:%S"),
parameters=[{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/grays/test-live-lots-raw.tsv"},
{"config_file":"/Users/patricktan/bitbucket/data/live-lots-etl/live-lots-etl-batch/config-dev-test.yaml",
"bucket":"wp-ahdata-sandbox",
"prefix":"dot-scrapy/live-lots/Blackwell/20220325_180155_Blackwell-DS=420.tsv"}]
)
wait_for_flow_a = wait_for_flow_run(mapped_flow_run_ids, raise_final_state=True)
flow.run()k
Kevin Kho
04/01/2022, 8:48 PM
Looking again, I think the idempotency key is weird also because it’s a string so you might need to supply something of type
List[str]p
Patrick Tan
04/01/2022, 8:49 PMso the number of idempotency_key should be same as my parameters
k
Kevin Kho
04/01/2022, 8:49 PM
yes for sure and they need to be distinct
Kevin Kho
04/01/2022, 8:57 PM
wait_for_flow_run(mapped_flow_run_ids, raise_final_state=True)Kevin Kho
04/01/2022, 8:57 PM
Because that task just takes in 1 id
p
Patrick Tan
04/01/2022, 9:15 PMThanks for all your help, not quite done but making progress. Will reach out next week if there is still issue. Have a good weekend
👍 1
9 Views