For no obvious reason, the storage used by my organization’s Nexus Repository Manager shrank from ~5.2TB to ~1.1 TB in the first week of June. No user impact was observed (or we would have detected it earlier), and no intentional changes were made to the configuration. No Nexus updates were installed, etc.
We have a large number of proxy repositories which we use to capture packages which may be needed to rebuild other items in a controlled manner, but we also store internally produced artifacts in our hosted repositories. All of our automatic Tasks occur on a frequency no longer than weekly, so an existing “cleanup” job wouldn’t be expected to find 80% of our storage usage to be eligible for removal.
Has anybody else seen anything like this? I have found some scripts to get the blobstore usage on a repository level, but I’m not familiar enough with Groovy scripting (or the internal Nexus storage) to quickly tweak such scripts to figure out what, if any, packages are present in a restored-from-back-up instance but not present in our production instance.
It seems that my options are to figure out how to tweak the script and then manually compare repository info output, or to basically click through every repository in both production and backup instances to figure out what (if any) packages are present in one but not the other. (Obviously newly-created packages in Production would be fine.)
Thoughts? We are in process of establishing a support relationship, but in the meanwhile…