Druid 0.8.3 RC3

Happy New Year to everyone!
We’re pleased to announce the latest Druid 0.8.3 release candidate – 0.8.3 RC3!

Wait, we skipped rc2, didn’t we? Well, we found some last minute bugs with restorable tasks right after releasing rc2, so we figured we’d wait for that.

You can download it here:

http://static.druid.io/artifacts/releases/druid-0.8.3-rc3-bin.tar.gz

Draft Release notes for the latest RC candidate are available at:

https://github.com/druid-io/druid/issues/2044

Please file GitHub issues if you find any bugs.

Thanks to all of you who contributed issues, docs, and code!

Can we update the webpage with this RC?

I’ve tried 0.8.3 RC4 in http://druid.io/downloads.html, but seeing complaints while starting a node about s3-extension not in Maven repo, might be something worth looking into. RC3 works fine for me.

As this is likely the first place anyone lands on 0.8.3, if you see the following error while indexing at realtime using tranquility, check the release note out: https://github.com/druid-io/druid/issues/2044 about setting `druid.selectors.coordinator.serviceName in common config. This thread is also helpful (https://groups.google.com/forum/m/#!topic/druid-user/-p5ecqYSYjY)

`2016-01-14T05:32:32,685 INFO [coordinator_handoff_scheduled_0] io.druid.segment.realtime.plumber.CoordinatorBasedSegmentHandoffNotifier - Still waiting for Handoff for Segments : [[SegmentDescriptor{interval=2016-01-14T03:00:00.000Z/2016-01-14T04:00:00.000Z, version=‘2016-01-14T03:06:28.156Z’, partitionNumber=15}]]
2016-01-14T05:33:32,685 ERROR [coordinator_handoff_scheduled_0] io.druid.curator.discovery.ServerDiscoverySelector - No server instance found
2016-01-14T05:33:32,685 ERROR [coordinator_handoff_scheduled_0] io.druid.segment.realtime.plumber.CoordinatorBasedSegmentHandoffNotifier - Exception while checking handoff for dataSource[test_10] Segment[SegmentDescriptor{interval=2016-01-14T03:00:00.000Z/2016-01-14T04:00:00.000Z, version=‘2016-01-14T03:06:28.156Z’, partitionNumber=15}], Will try again after [60000]secs
com.metamx.common.ISE: Cannot find instance of coordinator
at io.druid.client.coordinator.CoordinatorClient.baseUrl(CoordinatorClient.java:108) ~[druid-server-0.8.3-rc3.jar:0.8.3-rc3]
at io.druid.client.coordinator.CoordinatorClient.fetchServerView(CoordinatorClient.java:74) ~[druid-server-0.8.3-rc3.jar:0.8.3-rc3]
at io.druid.segment.realtime.plumber.CoordinatorBasedSegmentHandoffNotifier.checkForSegmentHandoffs(CoordinatorBasedSegmentHandoffNotifier.java:101) [druid-server-0.8.3-rc3.jar:0.8.3-rc3]
at io.druid.segment.realtime.plumber.CoordinatorBasedSegmentHandoffNotifier$1.run(CoordinatorBasedSegmentHandoffNotifier.java:86) [druid-server-0.8.3-rc3.jar:0.8.3-rc3]
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) [?:1.7.0_91]
at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:304) [?:1.7.0_91]
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:178) [?:1.7.0_91]
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293) [?:1.7.0_91]
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) [?:1.7.0_91]
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [?:1.7.0_91]

BTW, strongly recommend this release if you are experiencing slow merge before handing off segments. Thanks to this optimization https://github.com/druid-io/druid/pull/1960, previous IndexMerger took ~45minutes and now it’s ~1.5 minutes. This is amazing. Awesome work!

Thanks Shuai. There’s a bunch more optimizations that are going in as part of 0.9.0 including a major improvement here: https://github.com/druid-io/druid/pull/2138