Could not allocate segment for row with timestamp

[Arpan Khagram]

Was getting below error for my kafka indexing supervisors. Tried to search the forum but got different answers on all the topics. What this is about and what could be wrong ?

io.druid.java.util.common.ISE: Could not allocate segment for row with timestamp[2017-04-25T07:45:00.000Z]

[pja...@yahoo-inc.com]

You should check overlord logs around same time the exception happened in the task log to see why overlord could not allocate a new segment.

[Arpan Khagram]

Hi, This is what i found in overlord logs.

2017-04-25T12:45:15,203 INFO [qtp520354720-166] io.druid.indexing.common.actions.LocalTaskActionClient - Performing action for task[i

ndex_kafka_DSLAM_23c194dfbc16a9c_amaaheki]: SegmentAllocateAction{dataSource='DSLAM', timestamp=2017-04-25T07:45:00.000Z, queryGranul

arity=NoneGranularity, preferredSegmentGranularity={type=period, period=PT1H, timeZone=UTC, origin=null}, sequenceName='index_kafka_D

SLAM_23c194dfbc16a9c_1', previousSegmentId='null'}

2017-04-25T12:45:15,226 INFO [qtp520354720-166] io.druid.indexing.overlord.TaskLockbox - Added task[index_kafka_DSLAM_23c194dfbc16a9c

_amaaheki] to TaskLock[index_kafka_DSLAM]

2017-04-25T12:45:15,226 INFO [qtp520354720-166] io.druid.indexing.overlord.MetadataTaskStorage - Adding lock on interval[2017-04-25T0

7:00:00.000Z/2017-04-25T08:00:00.000Z] version[2017-04-25T11:15:16.118Z] for task: index_kafka_DSLAM_23c194dfbc16a9c_amaaheki

2017-04-25T12:45:15,237 WARN [qtp520354720-166] io.druid.metadata.IndexerSQLMetadataStorageCoordinator - Cannot use existing pending segment [DSLAM_2017-04-25T06:00:00.000Z_2017-04-25T07:00:00.000Z_2017-04-25T10:17:00.299Z_9] for sequence[index_kafka_DSLAM_23c194dfbc16a9c_1] (previous = []) in DB, does not match requested interval[2017-04-25T07:00:00.000Z/2017-04-25T08:00:00.000Z]