For some reason, we are getting non-terminal statuses when summarizing. It's possible this is a timing issue, so we'll naïvely attempt to get the stack description again until we receive a response with a terminal status.
By the time we make this call, we already know that we'll be in a terminal status because the watcher will only stop once we've seen a terminal status. This is why the logging will always have something like:
09:46:01 GMT-07:00 Thu Jun 20 2024 16:45:57 UPDATE_COMPLETE AWS::IAM::Role
09:46:01 GMT-07:00 CloudwatchLogRole (17s)
09:46:01 GMT-07:00 - 39 seconds elapsed total. 21 since last event.
09:46:06 GMT-07:00 Thu Jun 20 2024 16:46:00 UPDATE_COMPLETE_CLEANUP_IN_PROGRESS AWS::CloudFormation::Stack
09:46:06 GMT-07:00 eks-etc-titan-production
09:46:06 GMT-07:00 Thu Jun 20 2024 16:46:02 UPDATE_COMPLETE AWS::CloudFormation::Stack
09:46:06 GMT-07:00 eks-etc-titan-production (36s)
where the last line will be UPDATE_COMPLETE, but iidy will summarize with
09:46:08 GMT-07:00 Current Stack Status: UPDATE_COMPLETE_CLEANUP_IN_PROGRESS
which is clearly not the status from the last line.
@tavisrudd Although it did resolve the immediate issue, there was some concern about considering UPDATE_COMPLETE_CLEANUP_IN_PROGRESS a final state, so we're actually going with this approach.
For some reason, we are getting non-terminal statuses when summarizing. It's possible this is a timing issue, so we'll naïvely attempt to get the stack description again until we receive a response with a terminal status.
By the time we make this call, we already know that we'll be in a terminal status because the watcher will only stop once we've seen a terminal status. This is why the logging will always have something like:
where the last line will be
UPDATE_COMPLETE
, but iidy will summarize withwhich is clearly not the status from the last line.