[Bug]: aws_msk_replicator not tracking correct state of detect_and_copy_new_topics
#35964
Labels
bug
Addresses a defect in current functionality.
service/kafka
Issues and PRs that pertain to the kafka service.
Terraform Core Version
1.5.7
AWS Provider Version
5.38.0
Affected Resource(s)
Expected Behavior
Nothing. The state of the resource's
detect_and_copy_new_topics
attribute is alreadytrue
Actual Behavior
The plan shows that the provider wants to change that field from
false
totrue
. The state file also showsfalse
, though the resource itself @ AWS shows true. Attempting to change it from its current "true" to Terraform's "true" generates AWS error message:State at AWS:
State in
terraform show
:State after
state rm
andimport
is the same:Relevant Error/Panic Output Snippet
│ Error: updating MSK Replicator (arn:aws:kafka:us-east-1:179874059503:replicator/common-green-us-west-2-us-east-1/7e7885e6-0453-4623-9990-84759ba89d17-4): operation error Kafka: UpdateReplicationInfo, https response error StatusCode: 400, RequestID: 9f6ae9f2-d565-41f8-a4b1-1ef797d657a2, BadRequestException: The specified replication info is identical to the current value. Specify a different value, then try again.
Terraform Configuration Files
I'll try to do this later in the day; I'll want to extract a single replicator and single cluster pair to replicate with. This particular root module builds 5 MSK clusters & 4 replicators, so it's a bit big for a test case.
Steps to Reproduce
Debug Output
See https://gist.github.com/cobbr2/da1e354a303b6c43705108e8c5c382f3
Panic Output
No response
Important Factoids
These replicators have been destroyed & re-created dozens of times; only once we started setting this parameter to
true
did we run into this problem. We never used the default settings (we can't, we use MSK Serverless so the..._copy_acl...
stuff has to be set tofalse
).References
No response
Would you like to implement a fix?
None
The text was updated successfully, but these errors were encountered: