dang-stripe
dang-stripe
We've observed `select` and `select distinct` queries with no aggregation and a small limit clause can take a long time to process. We think that the broker is waiting for...
It'd be great to have an API to check rebalance status programmatically so we can automate rebalances. Currently the API always returns in-progress and requires checking the controller logs for...
We had a few segments that were not assigned to any servers due to metadata push happening when the controller was shutting down while it was getting replaced. The segments...
### Describe the bug We are using helix 1.3.1 from the pinot 1.2 release. We've observed a bug in our Pinot clusters where an expensive operation like rebalancing or heavy...
Follow up from https://github.com/apache/helix/issues/2951 which provides more detail. We performed a rebalance on an upsert table using low-disk mode that led to high GC on a server and the server...
This is a continuation of this issue: https://github.com/apache/pinot/issues/11894 ## Background When rebalancing upsert tables, we are noticing that nearly all segments are moving to a new host with `minimizeDataMovement` on....
We are testing some multistage queries with joins and adaptive routing turned on on a table using 3 replicas, 30 servers each. We joined against another much smaller table 90...
Based on this error message: https://github.com/apache/pinot/blob/master/pinot-controller/src/main/resources/app/pages/Query.tsx#L401-L406, it seems like queries can fail when segments aren't reloaded after a schema is updated. How often can this happen in practice and are...