[Bug]: Memory consumption has increased since v4.67.0 #31722
Description
Terraform Core Version
1.4.5
AWS Provider Version
4.67.0, 5.0.1, 5.1.0
Affected Resource(s)
Seems not related to specific resources.
For example, memory usage increases even when only declaring provider without resources.
Expected Behavior
Memory usage is kept the same by updating the provider.
Actual Behavior
In our actual environment setting up GuardDuty and EventBus on all available regions (17) on 5 AWS accounts, max RSS during refresh was measured as:
/usr/bin/time -v terraform plan -refresh-only| provider version | v4.66.1 | v4.67.0 | v5.0.1 |
|---|---|---|---|
| Max RSS | 2,645M | 3,654M | 3,634M |
| Increase | - | +38% | +37% |
Relevant Error/Panic Output Snippet
No response
Terraform Configuration Files
https://github.com/at-wat/terraform-provider-aws-v4.67-memory-usage/
Steps to Reproduce
See https://github.com/at-wat/terraform-provider-aws-v4.67-memory-usage/blob/main/README.md
This measures max RSS when refreshing aws_iam_policy_document data resources on 17 regions.
It results in a similar increase rate to our actual infrastructure code.
Debug Output
No response
Panic Output
No response
Important Factoids
No response
References
No response
Would you like to implement a fix?
None