Hello,
We are on a very old version of Rancher. Rancher 2.3.2. We run at large scale in the sense that we have 9 clusters, which each have many namespaces and run thousands of pods at a time. These are EKS clusters that we imported into Rancher years ago.
We have observed that Rancher slowly accumulates memory over time up to the point where it reaches 150GB of memory each week before we restart it and it goes back down to 10-15 GB of memory or so.
We know in our research that our version has a memory leak. Regardless, we don’t understand what Rancher reads into memory over time. For example, we don’t know if it holds data from the clusters it is managing such as pod related data in memory or if it holds connection data in memory or User related data etc.
In short, we don’t know what it holds in memory and are hoping someone could help shed some light on this.
Thanks for any help you can provide.