Druid Historical process fails to restart

Hello,

We are using Druid 0.13-incubating. A running historical process when stopped & then restarted, fails to bootstrap. And see a number of such log lines, followed by no logs at all.

{“msg”:“Allocating new littleEndByteBuf[81]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.947Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[82]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[83]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[85]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[86]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[84]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[87]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[88]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[89]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[90]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:47:37.948Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[92]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:48:04.239Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[91]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:48:04.239Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[93]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:48:04.239Z”,“lvlname”:“INFO”}

{“msg”:“Allocating new littleEndByteBuf[94]”,“name”:“org.apache.druid.segment.CompressedPools”,“lvlno”:“0”,“ts”:“2020-09-30T17:48:04.239Z”,“lvlname”:“INFO”}

Druid historical configs :

druid.query.groupBy.maxMergingDictionarySize=268435456

druid.server.http.numThreads=50

druid.processing.buffer.sizeBytes=2146435072

druid.processing.numThreads=23

druid.segmentCache.locations=[{“path”: “/data/druid/segment-cache”,“maxSize”: 1503238553600 }]

druid.server.maxSize=2040109465600

druid.historical.cache.useCache=true

druid.historical.cache.populateCache=true

druid.cache.type=caffeine

druid.cache.sizeInBytes=2147483647

druid.query.groupBy.maxOnDiskStorage=2146435072

We are running this historical on an : r5.12xlarge

This process finally came up after 30 mins, which seems like a really long time.

that’s a really old version of druid. you can download the newest version 0.19 at https://druid.apache.org/downloads.html