# AWS provider version 3.55

**URL:** <https://discuss.hashicorp.com/t/aws-provider-version-3-55/32055>\
**Category:** AWS\
**Created:** [November 17, 2021, 5:42pm UTC](https://discuss.hashicorp.com/t/aws-provider-version-3-55/32055 "2021-11-17T17:42:55Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![yshnayder](https://sea2.discourse-cdn.com/hashicorp/user_avatar/discuss.hashicorp.com/yshnayder/32/18564_2.png) [@yshnayder](https://discuss.hashicorp.com/u/yshnayder)\
**Post date:** [November 17, 2021, 5:42pm UTC](https://discuss.hashicorp.com/t/aws-provider-version-3-55/32055/1 "2021-11-17T17:42:55Z")

</div>

Hello,  
with the AWS provider migration from prior to 3.55 to 3.55 (where new DMS endpoint parameters were introduced) or later version, the existing DMS endpoints are not updated properly or not updated at all. The issue mostly applies to the `extra_connection_attributes` property. We execute `terraform apply` and it finishes successfully but when you check the actual configuration in AWS console you can see that it doesn’t match the one in tf file. If we issue another `terraform apply`, it shows the difference and you can apply the configuration again but the changes are not applied either. You can do it over and over with the successful result reported by terraform and the incorrect configuration appearing on the resource.  
Any thoughts?  
Thanks.