You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
could be improved in my opinion to avoid hitting rates limits:
It does all the requests in parallel (per function and per trigger), so x functions with y triggers will generate x*yparallel requests to the scaleway API.
It always deletes and recreate all triggers even if there are no changes.
The text was updated successfully, but these errors were encountered:
I'll try to improve the deploy trigger workflow, there is a complete delete/create of triggers each time because there are multiple sources of trigger CRUD (Scaleway Console, API, 3rd party tools etc...). The case an user wants to remove all of it's triggers even if created on Scaleway Console can be complicated.
I'll try to improve the deploy trigger workflow, there is a complete delete/create of triggers each time because there are multiple sources of trigger CRUD (Scaleway Console, API, 3rd party tools etc...).
The current behaviors (after serverless deploy the list of trigger exactly match the list specify in the config) seams correct for me. (Only the implementation needs a little improvement in my opinion)
The case an user wants to remove all of it's triggers even if created on Scaleway Console can be complicated.
If the config file specifies an empty array, it should delete all existing trigger. If the config doesn't specify schedule I think it should do the same.
When deploying ~20 functions with ~10 triggers each,
serverless deploy
hits rate limitsThe code handling deployment:
https://github.com/scaleway/serverless-scaleway-functions/blob/master/deploy/lib/deployTriggers.js#L19-L23
could be improved in my opinion to avoid hitting rates limits:
x
functions withy
triggers will generatex*y
parallel requests to the scaleway API.The text was updated successfully, but these errors were encountered: