Open erikgrinaker opened 2 years ago
We have marked this issue as stale because it has been inactive for 18 months. If this issue is still relevant, removing the stale label or adding a comment will keep it active. Otherwise, we'll close it in 10 days to keep the issue queue tidy. Thank you for your contribution to CockroachDB!
@kvoli Is this still relevant?
Still relevant. Write load is not accounted for directly. Store health (LSM) is only blocking rebalancing, not triggering.
The store rebalancer currently does not take into account replica/store write load or store health when making rebalancing decisions. This can lead to write hotspots that disrupt the entire cluster, and yet the rebalancer won't currently take action to resolve it. It needs to.
Related to #62168, #77604, #69364, #70911, #73714, and probably others.
Jira issue: CRDB-14643