> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/continuous-delivery/troubleshooting-and-resources/armory/general/spinnaker-does-not-properly-clean-up-old-artifacts-and-resources-when-redeploying.md).

# Spinnaker does not properly clean up old artifacts and resources when redeploying

### Issue <a href="#issue" id="issue"></a>

An organization may encounter an issue that Spinnaker does not delete old artifacts during a deployment, which are not part of the new deployment. This happens for Kubernetes deployments, which leads to a dirty cluster state and potential security risks e.g old istio rules being existent, and causes significant manual work needed to clean it up.

### Cause <a href="#cause" id="cause"></a>

This issue is caused by how OSS Spinnaker communicates with Kubernetes, there is currently no automatic way of determining what is and what isn't a used artifact or resource. This issue is being tracked on OSS Spinnaker <https://github.com/spinnaker/spinnaker/issues/4236>There is also related work going on in this issue <https://github.com/spinnaker/spinnaker/issues/4550>There are two sides to it. On the one side for singular pipelines that have a set number of resources, this can be handled with a delete resources stage as mentioned in the issue above.For multiple resource pipelines, it gets trickier as there currently isn't an ideal way of tracking those resources