Created
May 20, 2020 14:58
-
-
Save mffiedler/1ca6770f9c2e730fa9649016222e2624 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In general when things are left over from failed installs or failed destroy cluster, you need to go through resource by resource and look for your partial label (e.g. mffiedler). Oftentimes deleting the VPC will reap child resources, but to be thorough, go through (in this order): | |
S3: S3 bucket - this can be difficult to find. There could be two (one starts terraform and one image-registry) - use install log or cluster creation time to find them | |
EC2: Instances | |
EC2: Load Balancers (also search on the VPC IOD for ELBs that show up - there are sometimes "hidden" ELBs in the same VPC) | |
VPC: NAT Gateways (Delete 1-by-1, they take time to actually delete and can old up subsequent deletes, keep refreshing) | |
VPC: After waiting you can try to delete the VPC itself but it will likely complain about interfaces in use | |
VPC: If the VPC did not delete clean you likely have to go to the security group it complains about, try to delete it and then delete any resources it thinks are in use | |
VPC: Security group - search by label or security group ID and try to delete | |
EC2: Network interfaces | |
VPC: Delete VPC again - should succeed | |
IAM: users | |
Route 53: Hosted zones - have to delete record sets first, then hosted zones. Get to record sets by clicking hosted zones | |
For extreme cases like Walid's where the VPC is deleted but leaked resources may exist, also check: | |
VPC: Elastic IPs | |
VPC: Subnets | |
VPC: Endpoints | |
VPC: Security Groups | |
VPC: Route Tables | |
That's about as clean as you can get. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment