I'm curious why I'm seeing this in my coordinator ...
# troubleshootingj
JRob
04/21/2026, 2:24 PMI'm curious why I'm seeing this in my coordinator logs and if it would cause any problems in moving segments (i.e. that the min > max)
Copy code
2026-04-21T14:23:18,690 INFO [Coordinator-Exec-HistoricalManagementDuties-0] org.apache.druid.server.coordinator.balancer.SegmentToMoveCalculator - Need to move [368] segments in tier[_default_tier] to attain balance. Allowed values are [min=200, max=100].✅ 1
JRob
04/21/2026, 2:29 PMNote that balance never seems to be attained. Unsure if related to this or something else. I'm also having other issues in this cluster since a DC outage. All Zookeepers, Coordinators, and Historicals have been restarted but still seeing issues.
JRob
04/21/2026, 3:30 PMRelated, I see logs like this on the data nodes:
Copy code
2026-04-16T22:35:42,559 ERROR [Master-PeonExec--0] org.apache.druid.server.coordinator.loading.HttpLoadQueuePeon - Server[<http://druid-data20:8083>] failed segment[datasource_2026-02-25T16:00:00.000Z_2026-02-25T17:00:00.000Z_2026-02-25T18:09:30.045Z_1] request[DROP] with cause [Cannot update a path[/druid/segments/druid-data20:8083/druid-data20:8083_historical__default_tier_2026-04-16T19:11:28.073Z_7fc81e53461c41f7977eab3b4d59a49c182] that hasn't been announced!].JRob
04/21/2026, 3:33 PM Copy code
2026-04-17T14:13:16,537 ERROR [qtp2118829464-285] org.apache.druid.server.coordination.SegmentLoadDropHandler - Failed to remove segment: {exceptionType=org.apache.druid.java.util.common.ISE, exceptionMessage=Cannot update a path[/druid/segments/druid-data03:8083/druid-data03:8083_historical__default_tier_2026-04-17T13:53:38.831Z_3baa39b68f7347baa9a92262378fb1fc177] that hasn't been announced!, class=org.apache.druid.server.coordination.SegmentLoadDropHandler, segment=DataSegment{binaryVersion=9, id=datasource_2026-04-04T00:00:00.000Z_2026-04-05T00:00:00.000Z_2026-04-04T00:00:00.377Z_23, loadSpec={type=>local, path=>/data/deep-storage/segments/datasource/2026-04-04T00:00:00.000Z_2026-04-05T00:00:00.000Z/2026-04-04T00:00:00.377Z/23/0dfc0091-c62c-480c-94ac-bd2f46e0b037/index/}, dimensions=[omitted], metrics=[count], shardSpec=NumberedShardSpec{partitionNum=23, partitions=0}, lastCompactionState=null, size=7929}}JRob
04/21/2026, 4:38 PMI've concluded that the SegmentToMoveCalculator stuff appears to be normal operation. It always hovers around ~350 in my cluster. The errors are not related.
2 Views