Hi, thanks for getting in touch with this question.
Before I give you any advice, I just want to clarify what you are asking. Firstly, are you running a health check as part of a machine policy every minute (or is it part of a deployment). Also, can you let me know why the rows are an issue for you and which table and rows you are referring to.
In general, deleting from the database can have unintended consequences, and if there is a setting that helps achieve what you require I'll help you set it up.
We are utilizing Octopus to execute PowerShell scripts that are a part of a deployment being executed every minute. We wrote a task scheduler that handle this. The rows that are of a concern are in the deployment database. A line is being written every deployment. In this case we are deploying every minute and writing approximate 1440 lines per day. We don’t need the logs for the deployment of the health script and would like to eliminate them from that table. Is there a way to set it up to not write it out to the database?
If you want to delete the whole deployment and it's associated logs from Octopus. The best way is to use the Deployment delete endpoint in the Web API. Check out the swagger docs at [YourOctopusServerURL]/swaggerui/.
We've also got a new feature coming for recurring scheduled deployments, which might be of interest to you. You can read about it here. We are planning to release it in our March release.