Skip to content

gracefully handle 500 errors on server requests #118

@adobeDan

Description

@adobeDan

Things happen sometimes that cause servers to go boom, and user-sync should not choke when that happens (as documented in #114 and #116). Given that most of user-sync's actions are idempotent, probably it should try to retry a batch of actions when they fail, but at worst it should just log the failure and move on as if the operation succeeded. Maybe we even need a setting for what should happen in this case.

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions