@Jason Bosco I am using Docker in 3 different nodes with Public IPS. I am building a 3-node cluster. One of the nodes stopped when I was doing the incersion of the data. The data continued to be inserted in node 2 and 3. I manage to raise the node 1, and now the memory used in node 1, 2, 3 are different.

@Nelson Moncada On startup, the recovering node will reach out to the other nodes in the cluster to fetch the latest snapshot, and then rebuild the index in memory. This will take time, and that's what causes the difference in memory consumption

Ok thk

Yup, same reason. It's re-building the index

Ok