Tentacle retention policy during deployment is slow

[shared.services]

We currently have 2022.2.7934 installed and noticed that during each deployment, it took a lot of time to run the retention policies on Tentacle even if it doesn’t do lots of work.

The problem is as you may see from the screenshot, it runs the retention policy for each step in the process. Even though each step only took 3-4 seconds, it adds up to over 1 mins even if most of the time it doesn’t clean anything. Can this be at least parallelized or optimized so that it won’t waste too much time for each deployment? Thanks

[paraic.oceallaigh]

Hi @shared.services
Thanks for sending through this query.

Looking at the figures you supplied in the screenshot, I would say those timings are not extreme as connecting to a Tentacle, scanning through multiple folders and comparing them might be expected to take a second or more depending on your environment.

That is not to say its not an issue if you have very large folder structures and I could see that even one second per folder could be an issue for large scale deploys.

To be honest we don’t have any elegant solution for this at present bar using something like Lifecycles where conditions can be set per project on individual retention policies:
https://octopus.com/docs/administration/retention-policies#retention-policies-and-lifecycle-phases

You could also, in a more inelegant fashion select custom install folders as an option - these are never scanned for retention policies, but instead typically a step is included in the process to delete artefacts:
https://octopus.com/docs/projects/steps/configuration-features/custom-installation-directory

You can also submit a suggestion on our Roadmap page where we ask customers for their suggestions for upcoming releases:
https://roadmap.octopus.com/tabs/2-planned. (just click the Submit idea on the top right).

Sorry we can’t offer a better solution at this time.

Kind regards,
Paraic