High blob count in group repositories

Hello!
I am using Sonatype Nexus Repository Manager
OSS 3.49.0-02 with embedded OrientDB. I started to analyze the potential cause of repetitive application errors that occur even after scaling the host to 32GB RAM, with the settings compliant with nexus requirments document:

-Xms6G
-Xmx6G
-XX:MaxDirectMemorySize=15530M

It turned out that the sum of blobs from individual repositories exceeded 147,000, with the allowable database limit of 100,000 objects.
So I started cleaning the repositories from obsolete objects, but it turned out that I can’t clean the group repositories. Moreover, the number of objects and their size in the group significantly exceeded the sum and size of objects in the repositories it grouped.
Using the npm repository as an example:


npm is a repository grouping the npm-hosted and npm-proxy repositories

Cleaning and compacting individual repositories did not affect the number and size of the grouping repository.
Therefore, I have a question: Why do the number of objects in groups exceed the number of objects in grouped repositories, and how can I purge grouping repositories?

There are some known edge cases with the counts and the totals which cause them to be inaccurate. There’s a ticket that exists: Loading.... My understanding is when that ticket is complete known edge cases will be resolved and there will be a task which allows you to do what you ask and “reset” the counts to the correct values. The task would allow us to see any further anomalies. My suggestion would be to wait until that work is done and see if you still have outstanding issues. Sorry that’s not immediately helpful but at least providing information.