If I am launching my meltano runs in my orchestrat...
# troubleshooting
a
If I am launching my meltano runs in my orchestrator using a job command (i.e.
meltano run google_analytics
) is there a way to configure my job or task in the yaml to use
--force
?
Copy code
- name: google_analytics
  tasks:
  - tap-google-analytics target-postgres
Adding
--force
to the block returns
Error: Block --force not found
v
meltano run --force google_analytics
not
meltano run google_analytics --force
e
Yeah, this is not supported at the moment
Copy code
- name: google_analytics
  tasks:
  - --force tap-google-analytics target-postgres
I'm curious why you'd prefer to hard-set the
--force
flag.
a
@Edgar Ramírez (Arch.dev) I run the jobs via Dagster in Azure App Service. If I cancel the job in dagster, meltano doesn't exit gracefully, and sometimes thinks there is another run still ongoing according to system db. That prevents me from restarting the run via dagster for a little while. Unfortunately (AFAIK) I have no way to alter the meltano command being run by dagster, it's preset in my definition. So hard-coding
--force
seemed like the next best option.
v
If I cancel the job in dagster, meltano doesn't exit gracefully, and sometimes thinks there is another run still ongoing according to system db
Is a fixable problem
Unfortunately (AFAIK) I have no way to alter the meltano command being run by dagster, it's preset in my definition.
Is fixable
So hard-coding
--force
seemed like the next best option.
+1
e
Do log a feature request 🙏 I think there's a few ways we could do this. A community PR recently added an env var for
--full-refresh
, so this might do something similar. The difficulty of hard-coding it in the job definition is that
meltano run
can execute more than job so we decide how we reduce them. I gave it a rough pass in • https://github.com/meltano/meltano/tree/edgarrmondragon/feat/declarative-force-jobhttps://github.com/meltano/meltano/actions/runs/11849511341 but I'm not a fan of this approach