Bulk uploader not making equal shards

@Valdanito thanks for the reply.

cluster Info
Alpha Nodes: 7
Zero Nodes: 1
Shard Replica: 1
(Essentially 7 groups with no replication)

That was my understanding too and I confirm from the ratel!!! But, what I don’t understand is…

  • This cluster was stood up about 72 hrs ago
  • Based on the K8S logs, all along predicates are on the move on two Alpha nodes (see attached screen capture highlighted in red circle) which I don’t understand
  • Below is the typical log entry
   I0314 17:45:14.678083      20 schema.go:496] Setting schema for attr Encounter.timestamp: datetime, tokenizer: [year], directive: INDEX, count: false                                                                 ││ I0314 17:45:15.057332      20 predicate_move.go:48] Writing 335545 keys                                                                                                                                               ││ I0314 17:45:15.658652      20 log.go:34] L0 was stalled for 5.479s                                                                                                                                                    ││ I0314 17:45:15.721073      20 log.go:34] [0] [E] LOG Compact 0->5 (8, 55 -> 58 tables with 5 splits). [297058 297069 297075 297086 297092 297102 297110 297118 . 297066 297072 297079 297083 297089 297095 297099 297 ││ I0314 17:45:17.902581      20 schema.go:496] Setting schema for attr Encounter.timestamp: datetime, tokenizer: [year], directive: INDEX, count: false