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
We want to be able to distinguish these conditions, which can be due
to internal misconfiguration or external Cincinnati/network errors
[1]. The former can be fixed by cluster admins. The latter could go
either way.
I dropped the len(upstream) guard from checkForUpdate because there's
already an earlier guard in syncAvailableUpdates. The guard I'm
removing is from db150e6 (cvo: Perform status updates in a single
thread, 2018-11-03, #45). The covering guard is from the later
286641d (api: Update to objects from openshift/api, 2018-11-15, #55).
Personally, I'd rather have GetUpdates return an *Error, so we could
dispense with the cast and unused Unknown-reason fallback. But
Abhinav wanted the explicit cast in return for a more familiar error
type [2].
[1]: https://bugzilla.redhat.com/show_bug.cgi?id=1685338
[2]: #268 (comment)
0 commit comments