19913d85-f662-4518-a9a1-7adc1793dfd4.png){kind=small}
I have a setup where sometimes I need to re-execut...
# ask-communityj
Jared Wolff
05/03/2021, 7:15 PMI have a setup where sometimes I need to re-execute a run when input data is adjusted or added. I'd like to maintain the _run_key_ and programmatically trigger the re-execute action on the dashboard interface. I see that there's a reexecute() API function, which takes a _run_id_, but the only API I ran across to look up a _run_id_ is listed as internal and subject to change. Is there a better approach I can take?
m
max
05/04/2021, 12:20 AMwhich API were you thinking of? (for looking up a run_id)
j
Jared Wolff
05/04/2021, 12:26 AMThere's a
get_runs()Jared Wolff
05/04/2021, 12:27 AMMy _run_key_ is tied to the data, so I can use that as my filter
m
max
05/04/2021, 3:51 AMare you kicking off your runs programmatically or in Dagit?
j
Jared Wolff
05/04/2021, 4:40 AMTo generate the initial run, I have a sensor which calls RunRequest(). However, sometimes I need to re-execute some of the runs.
m
max
05/04/2021, 5:35 AMgotcha. I think you might be better off calling methods on the
DagsterInstancemax
05/04/2021, 5:35 AMthis is less likely to change than the backing storages
j
Jared Wolff
05/04/2021, 1:17 PMIs there an API for that?
Jared Wolff
05/12/2021, 2:03 AMI was able to pull most of the data I need from get_runs(), but the PipelineRun class doesn't have the startTime and endTime of the pipeline. I see that I do have a snapshot_id, which it appears I can use to look that info up - but I don't see a function to return it.