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
a new exception type SkipPartitionAPIError is added, which tap developers can raise to signify "this partition cannot be synced, but other partitions might, so continue with the next one".
unless that exception is raised, the tap's behaviour is unchanged,
in particular, if FatalAPIError is raised, the tap will stop immediately.
Typical use case for this exception would be a server-side error on the API part which cannot be resolved with the tap, other than by retrying at a later date. We are seeing quite a lot of these on the github API, with no clear pattern or cause. In most cases, retrying within a few seconds is enough, but sometimes it is not. In that case, the entire tap run is currently cancelled, which seems a fairly high price to pay.
Obviously happy to have feedback on this and adjust to make it as useful as possible.
This has been marked as stale because it is unassigned, and has not had recent activity. It will be closed after 21 days if no further activity occurs. If this should never go stale, please add the evergreen label, or request that it be added.
Merges skip-partition-on-error -> main
Migrated from GitLab: https://gitlab.com/meltano/sdk/-/merge_requests/258
This PR is a proposal to address https://gitlab.com/meltano/sdk/-/issues/282
The change proposed here is fully opt-in:
SkipPartitionAPIError
is added, which tap developers can raise to signify "this partition cannot be synced, but other partitions might, so continue with the next one".FatalAPIError
is raised, the tap will stop immediately.Typical use case for this exception would be a server-side error on the API part which cannot be resolved with the tap, other than by retrying at a later date. We are seeing quite a lot of these on the github API, with no clear pattern or cause. In most cases, retrying within a few seconds is enough, but sometimes it is not. In that case, the entire tap run is currently cancelled, which seems a fairly high price to pay.
Obviously happy to have feedback on this and adjust to make it as useful as possible.
Closes #282
The text was updated successfully, but these errors were encountered: