Suspicious Volume Usage on Workers

As a Product Manger at Pivotal I’m often called on to help with our customer’s Concourse-related issues. I recently spent some time hunting down an issue around suspiciously high volume usage on Concourse workers. It was an interesting problem that I wanted to share with the broader Concourse community.

This is a companion discussion topic for the original entry at

In the example above, not clear on how pivnet reports a new version on basis of metadata. If you could throw some light on that would be helpful for us to relate. We face a similar issue of concourse nodes getting filled. But ours is pas backup pipeline using bbr pas backup.

Thanks and Regards,