Commit Graph

679 Commits

Author SHA1 Message Date
Gian Merlino a87db7f353
Add HashJoinSegment, a virtual segment for joins. (#9111)
* Add HashJoinSegment, a virtual segment for joins.

An initial step towards #8728. This patch adds enough functionality to implement a joining
cursor on top of a normal datasource. It does not include enough to actually do a query. For
that, future patches will need to wire this low-level functionality into the query language.

* Fixups.

* Fix missing format argument.

* Various tests and minor improvements.

* Changes.

* Remove or add tests for unused stuff.

* Fix up package locations.
2020-01-16 13:14:20 -08:00
Chi Cao Minh 1fd05bef9a Add jackson-mapper-asl for hdfs-storage extension (#9178)
Previously jackson-mapper-asl was excluded to remove a security
vulnerability; however, it is required for functionality (e.g.,
org.apache.hadoop.security.token.delegation.web.DelegationTokenAuthenticator).
2020-01-14 09:50:45 -08:00
Atul Mohan ea51bc45bf Fix nullhandling in tests (#9119) 2020-01-12 20:19:12 -08:00
Clint Wylie 85219ece13 fix null handling for arithmetic post aggregator comparator (#9159)
* fix null handling for arithmetic postagg comparator, add test for comparator for min/max/quantile postaggs in histogram ext

* fix
2020-01-10 13:49:19 -08:00
Jihoon Son e27a1e8604
Fix handling nullable writableComparable in OrcStructConverter (#9138)
* Handle nullable writableComparable in OrcStructConverter

* add missing dependency
2020-01-08 13:40:24 -08:00
Clint Wylie f540216931 fix InputFormat serde issue with SeekableStream based supervisors (#9136) 2020-01-07 16:18:54 -06:00
Clint Wylie 7af85250cb null handling for doubles sketch and array of doubles sketch aggs (#9112)
* doubles sketch and array of doubles sketch aggs now skip rows with nulls in sql compatible null handling mode

* formatting
2020-01-07 14:15:32 -06:00
Suneet Saldanha bdd0d0d8a5 Add avro dependency to parquet extension (#9124)
* Add avro dependency to parquet extension

If the parquet extension is loaded and an ingestionSpec uses the older format
specifying a 'parser' instead of using an 'inputFormat' the job fails
with the following error

java.lang.TypeNotPresentException: Type org.apache.avro.generic.GenericRecord not present

This change removes the exclusion of the avro package so that the missing
class can be found.

* Address review comments and add dependency version
2020-01-03 20:11:13 -06:00
Jonathan Wei aa539177ec De-incubation cleanup in code, docs, packaging (#9108)
* De-incubation cleanup in code, docs, packaging

* remove unused docs script
2020-01-03 12:33:19 -05:00
Jonathan Wei 4e8368a5d9 Set version to 0.18.0-SNAPSHOT (#9109) 2020-01-02 17:55:10 -05:00
Gian Merlino 18eb456fe6
S3: Improvements to prefix listing (including fix for an infinite loop) (#9098)
* S3: Improvements to prefix listing (including fix for an infinite loop)

1) Fixes #9097, an infinite loop that occurs when more than one batch
   of objects is retrieved during a prefix listing.

2) Removes the Access Denied fallback code added in #4444. I don't think
   the behavior is reasonable: its purpose is to fall back from a prefix
   listing to a single-object access, but it's only activated when the
   end user supplied a prefix, so it would be better to simply fail, so
   the end user knows that their request for a prefix-based load is not
   going to work. Presumably the end user can switch from supplying
   'prefixes' to supplying 'uris' if desired.

3) Filters out directory placeholders when walking prefixes.

4) Splits LazyObjectSummariesIterator into its own class and adds tests.

* Adjust S3InputSourceTest.

* Changes from review.

* Include hamcrest-core.
2019-12-31 19:06:49 -05:00
Chi Cao Minh 513bb1f6da Get proper Kinesis index task AWS credentials (#9082)
Previously, the configured S3 credentials would be used instead of the
ones configured for Kinesis for Kinesis index tasks.
2019-12-20 19:35:05 -08:00
Jihoon Son 66056b2826
Using annotation to distinguish Hadoop Configuration in each module (#9013)
* Multibinding for NodeRole

* Fix endpoints

* fix doc

* fix test

* Using annotation to distinguish Hadoop Configuration in each module
2019-12-11 17:30:44 -08:00
Jonathan Wei 8af41d7cd0 Update version to 0.18.0-incubating-SNAPSHOT (#9009) 2019-12-11 14:04:03 -08:00
Chi Cao Minh 3de7ab8523 DataSketches jars in core (#9003)
Having DataSketches jars in core will allow potential improvements, for
example:
- Provide an alternative implementation of HLL:
  https://datasketches.github.io/docs/HLL/HllSketchVsDruidHyperLogLogCollector.html
- Range partitioning for native parallel batch indexing without having
  the user load extensions on the classpath

Dev mailing list discussion:
https://lists.apache.org/thread.html/301410d71ff799cf616bf17c4ebcf9999fc30829f5fa62909f403e6c%40%3Cdev.druid.apache.org%3E
2019-12-10 14:02:34 -08:00
Chi Cao Minh bab78fc80e Parallel indexing single dim partitions (#8925)
* Parallel indexing single dim partitions

Implements single dimension range partitioning for native parallel batch
indexing as described in #8769. This initial version requires the
druid-datasketches extension to be loaded.

The algorithm has 5 phases that are orchestrated by the supervisor in
`ParallelIndexSupervisorTask#runRangePartitionMultiPhaseParallel()`.
These phases and the main classes involved are described below:

1) In parallel, determine the distribution of dimension values for each
   input source split.

   `PartialDimensionDistributionTask` uses `StringSketch` to generate
   the approximate distribution of dimension values for each input
   source split. If the rows are ungrouped,
   `PartialDimensionDistributionTask.UngroupedRowDimensionValueFilter`
   uses a Bloom filter to skip rows that would be grouped. The final
   distribution is sent back to the supervisor via
   `DimensionDistributionReport`.

2) The range partitions are determined.

   In `ParallelIndexSupervisorTask#determineAllRangePartitions()`, the
   supervisor uses `StringSketchMerger` to merge the individual
   `StringSketch`es created in the preceding phase. The merged sketch is
   then used to create the range partitions.

3) In parallel, generate partial range-partitioned segments.

   `PartialRangeSegmentGenerateTask` uses the range partitions
   determined in the preceding phase and
   `RangePartitionCachingLocalSegmentAllocator` to generate
   `SingleDimensionShardSpec`s.  The partition information is sent back
   to the supervisor via `GeneratedGenericPartitionsReport`.

4) The partial range segments are grouped.

   In `ParallelIndexSupervisorTask#groupGenericPartitionLocationsPerPartition()`,
   the supervisor creates the `PartialGenericSegmentMergeIOConfig`s
   necessary for the next phase.

5) In parallel, merge partial range-partitioned segments.

   `PartialGenericSegmentMergeTask` uses `GenericPartitionLocation` to
   retrieve the partial range-partitioned segments generated earlier and
   then merges and publishes them.

* Fix dependencies & forbidden apis

* Fixes for integration test

* Address review comments

* Fix docs, strict compile, sketch check, rollup check

* Fix first shard spec, partition serde, single subtask

* Fix first partition check in test

* Misc rewording/refactoring to address code review

* Fix doc link

* Split batch index integration test

* Do not run parallel-batch-index twice

* Adjust last partition

* Split ITParallelIndexTest to reduce runtime

* Rename test class

* Allow null values in range partitions

* Indicate which phase failed

* Improve asserts in tests
2019-12-09 23:05:49 -08:00
Roman Leventov 1c62987783
Add SelfDiscoveryResource; rename org.apache.druid.discovery.No… (#6702)
* Add SelfDiscoveryResource

* Rename org.apache.druid.discovery.NodeType to NodeRole. Refactor CuratorDruidNodeDiscoveryProvider. Make SelfDiscoveryResource to listen to updates only about a single node (itself).

* Extended docs

* Fix brace

* Remove redundant throws in Lifecycle.Handler.stop()

* Import order

* Remove unresolvable link

* Address comments

* tmp

* tmp

* Rollback docker changes

* Remove extra .sh files

* Move filter

* Fix SecurityResourceFilterTest
2019-12-08 18:47:58 +03:00
Chi Cao Minh af74acaa85 Address security vulnerabilities CVSS >= 7 (#8980)
* Address security vulnerabilities CVSS >= 7

Update dependencies to address security vulnerabilities with CVSS scores
of 7 or higher. A new Travis CI job is added to prevent new
high/critical security vulnerabilities from being added.

Updated dependencies:
- api-util 1.0.0 -> 1.0.3
- jackson 2.9.10 -> 2.10.1
- kafka 2.1.0 -> 2.1.1
- libthrift 0.10.0 -> 0.13.0
- protobuf 3.2.0 -> 3.11.0

The following high/critical security vulnerabilities are currently
suppressed (so that the new Travis CI job can be added now) and are left
as future work to fix:
- hibernate-validator:5.2.5
- jackson-mapper-asl:1.9.13
- libthrift:0.6.1
- netty:3.10.6
- nimbus-jose-jwt:4.41.1

* Rename EDL1 license file

* Fix inspection errors
2019-12-05 14:34:35 -08:00
Clint Wylie 5ecdf94d83
add 'prefixes' support to google input source (#8930)
* add prefixes support to google input source, making it symmetrical-ish with s3

* docs

* more better, and tests

* unused

* formatting

* javadoc

* dependencies

* oops

* review comments

* better javadoc
2019-12-04 21:01:10 -08:00
Clint Wylie b4efaa698b unexclude necessary jackson mapper-asl jars (#8977) 2019-12-02 17:01:11 -08:00
Chi Cao Minh 4b7e79a4e6 Exclude unneeded hadoop transitive dependencies (#8962)
* Exclude unneeded hadoop transitive dependencies

These dependencies are provided by core:
- com.squareup.okhttp:okhttp
- commons-beanutils:commons-beanutils
- org.apache.commons:commons-compress
- org.apache.zookepper:zookeeper

These dependencies are not needed and are excluded because they contain
security vulnerabilities:
- commons-beanutils:commons-beanutils-core
- org.codehaus.jackson:jackson-mapper-asl

* Simplify exclusions + separate unneeded/vulnerable

* Do not exclude jackson-mapper-asl
2019-12-02 16:08:21 -08:00
Clint Wylie 6997b167b1 add hdfs client dependency for native batch parquet when using hdfs (#8964) 2019-11-28 13:12:45 -08:00
Jonathan Wei 00ce18a0ea
Additional Kinesis resharding fixes (#8870)
* Additional Kinesis resharding fixes

* Address PR comments

* Remove unused method

* Adjust SegmentTransactionalInsertAction null handling

* Check for unchanged metadata on empty publish

* Add logs for empty publish

* Fix javadoc

* Clear offset when invalid endOffsets are seen

* Fix LGTM alert

* Fix build

* Add resharding note to Kinesis docs

* Checkstyle

* Spelling

* Address PR comments

* Checkstyle
2019-11-28 12:59:01 -08:00
Jihoon Son 86e8903523
Support orc format for native batch ingestion (#8950)
* Support orc format for native batch ingestion

* fix pom and remove wrong comment

* fix unnecessary condition check

* use flatMap back to handle exception properly

* move exceptionThrowingIterator to intermediateRowParsingReader

* runtime
2019-11-28 12:45:24 -08:00
jon-wei dfbc066163 Revert "[maven-release-plugin] prepare release druid-0.16.1-incubating-rc1"
This reverts commit a0f21d9b07.
2019-11-27 23:22:43 -08:00
jon-wei 0402ff85b8 Revert "[maven-release-plugin] prepare for next development iteration"
This reverts commit 8ffa71e7e6.
2019-11-27 23:22:32 -08:00
jon-wei 8ffa71e7e6 [maven-release-plugin] prepare for next development iteration 2019-11-27 23:18:48 -08:00
jon-wei a0f21d9b07 [maven-release-plugin] prepare release druid-0.16.1-incubating-rc1 2019-11-27 23:18:37 -08:00
Clint Wylie 4458113375
S3 input source (#8903)
* add s3 input source for native batch ingestion

* add docs

* fixes

* checkstyle

* lazy splits

* fixes and hella tests

* fix it

* re-use better iterator

* use key

* javadoc and checkstyle

* exception

* oops

* refactor to use S3Coords instead of URI

* remove unused code, add retrying stream to handle s3 stream

* remove unused parameter

* update to latest master

* use list of objects instead of object

* serde test

* refactor and such

* now with the ability to compile

* fix signature and javadocs

* fix conflicts yet again, fix S3 uri stuffs

* more tests, enforce uri for bucket

* javadoc

* oops

* abstract class instead of interface

* null or empty

* better error
2019-11-25 22:31:19 -08:00
Alexander Saydakov 4a9da3f3fc use the latest release of datasketches (#8647)
* use the latest release of datasketches

* added datasketches-memory dependency

* updated datasketches entries

* use datasketches-memory-1.2.0

* updated dependencies

* fixed tests
2019-11-25 19:45:51 -08:00
Clint Wylie cd31bcc093 un-exclude necessary parquet jackson dependencies instead of relying on curator (#8939) 2019-11-25 15:57:34 -08:00
Jihoon Son a2e6de4b16 Fix the potential race between SplittableInputSource.getNumSplits() and SplittableInputSource.createSplits() in TaskMonitor (#8924)
* Fix the potential race SplittableInputSource.getNumSplits() and SplittableInputSource.createSplits() in TaskMonitor

* Fix docs and javadoc

* Add unit tests for large or small estimated num splits

* add override
2019-11-23 01:38:08 -08:00
Gian Merlino e0eb85ace7 Add FileUtils.createTempDir() and enforce its usage. (#8932)
* Add FileUtils.createTempDir() and enforce its usage.

The purpose of this is to improve error messages. Previously, the error
message on a nonexistent or unwritable temp directory would be
"Failed to create directory within 10,000 attempts".

* Further updates.

* Another update.

* Remove commons-io from benchmark.

* Fix tests.
2019-11-22 19:48:49 -08:00
Clint Wylie 7250010388 add parquet support to native batch (#8883)
* add parquet support to native batch

* cleanup

* implement toJson for sampler support

* better binaryAsString test

* docs

* i hate spellcheck

* refactor toMap conversion so can be shared through flattenerMaker, default impls should be good enough for orc+avro, fixup for merge with latest

* add comment, fix some stuff

* adjustments

* fix accident

* tweaks
2019-11-22 10:49:16 -08:00
Jihoon Son 934547a215
RetryingInputEntity to retry on transient errors (#8923)
* RetryingInputEntity to retry on transient errors

* fix some javadoc and httpEntity

* Make it interface

* Javadoc for offset
2019-11-21 21:32:18 -08:00
Jonathan Wei dc6178d1f2 Upgrade Calcite to 1.21 (#8566)
* Upgrade Calcite to 1.21

* Checkstyle, test fix'

* Exclude calcite yaml deps, update license.yaml

* Add method for exception chain handling

* Checkstyle

* PR comments, Add outer limit context flag

* Revert project settings change

* Update subquery test comment

* Checkstyle fix

* Fix test in sql compat mode

* Fix test

* Fix dependency analysis

* Address PR comments

* Checkstyle

* Adjust testSelectStarFromSelectSingleColumnWithLimitDescending
2019-11-20 21:22:55 -08:00
Jihoon Son ac6d703814 Support inputFormat and inputSource for sampler (#8901)
* Support inputFormat and inputSource for sampler

* Cleanup javadocs and names

* fix style

* fix timed shutoff input source reader

* fix timed shutoff input source reader again

* tidy up timed shutoff reader

* unused imports

* fix tc
2019-11-20 14:51:25 -08:00
Surekha d628bebbd7 Make supervisor API similar to submit task API (#8810)
* accept spec or dataSchema, tuningConfig, ioConfig while submitting task json

* fix test

* update docs

* lgtm warning

* Add original constructor back to IndexTask to minimize changes

* fix indentation in docs

* Allow spec to be specified in supervisor schema

* undo IndexTask spec changes

* update docs

* Add Nullable and deprecated annotations

* remove deprecated configs from SeekableStreamSupervisorSpec

* remove nullable annotation
2019-11-20 10:04:41 -08:00
Clint Wylie 3fcaa1a61b
fix sql compatible null handling config work with runtime.properties (#8876)
* fix sql compatible null handling config work with runtime.properties

* fix npe

* fix tests

* add friendly error

* comment, and friendlier still

* fix compile

* fix from merges
2019-11-20 03:55:29 -08:00
Chi Cao Minh 4ae6466ae2 HDFS input source (#8899)
* HDFS input source

Add support for using HDFS as an input source. In this version, commas
or globs are not supported in HDFS paths.

* Fix forbidden api

* Address review comments
2019-11-19 22:19:39 -08:00
Clint Wylie 074a45219d add google cloud storage InputSource for native batch (#8907)
* add google cloud storage InputSource for native batch

* rename

* checkstyle

* fix

* fix spelling

* review comments
2019-11-19 19:49:43 -08:00
Rye d0913475b7 sampler returns nulls in CSV (#8871)
* sampler returns nulls in CSV

* fixed kafka sampler test

* fix Kinesis test

* sql compatibility fix

* remove null to empty string conversion, use null

* fix sql compatibility
2019-11-19 13:59:44 -08:00
Gian Merlino c44452f0c1 Tidy up lifecycle, query, and ingestion logging. (#8889)
* Tidy up lifecycle, query, and ingestion logging.

The goal of this patch is to improve the clarity and usefulness of
Druid's logging for cluster operators. For more information, see
https://twitter.com/cowtowncoder/status/1195469299814555648.

Concretely, this patch does the following:

- Changes a lot of INFO logs to DEBUG, and DEBUG to TRACE, with the
  goal of reducing redundancy and improving clarity by avoiding
  showing rarely-useful log messages. This includes most "starting"
  and "stopping" messages, and most messages related to individual
  columns.
- Adds new log4j2 templates that show operators how to enabled DEBUG
  logging for certain important packages.
- Eliminate stack traces for query errors, unless log level is DEBUG
  or more. This is useful because query errors often indicate user
  error rather than system error, but dumping stack trace often gave
  operators the impression that there was a system failure.
- Adds task id to Appenderator, AppenderatorDriver thread names. In
  the default log4j2 configuration, this will put them in log lines
  as well. It's very useful if a user is using the Indexer, where
  multiple tasks run in the same JVM.
- More consistent terminology when it comes to "sequences" (sets of
  segments that are handed-off together by Kafka ingestion) and
  "offsets" (cursors in partitions). These terms had been confused in
  some log messages due to the fact that Kinesis calls offsets
  "sequence numbers".
- Replaces some ugly toString calls with either the JSONification or
  something more operator-accessible (like a URL or segment identifier,
  instead of JSON object representing the same).

* Adjustments.

* Adjust integration test.
2019-11-19 13:57:58 -08:00
Surekha cf6643eb9a add sequenceName and currentCheckPoint for backwards compatibility (#8864)
* add sequenceName and currentCheckPoint for backwards compatibility

* Add serde unit test in kafka

* fix checkstyle

* add hashcode

* update javadoc
2019-11-19 13:11:31 -08:00
Chi Cao Minh 8365bdf62a Address security vulnerabilities (#8878)
* Address security vulnerabilities

Security vulnerabilities addressed by upgrading 3rd party libs:

- Upgrade avro-ipc to 1.9.1
  - sonatype-2019-0115
- Upgrade caffeine to 2.8.0
  - sonatype-2019-0282
- Upgrade commons-beanutils to 1.9.4
  - CVE-2014-0114
- Upgrade commons-codec to 1.13
  - sonatype-2012-0050
- Upgrade commons-compress to 1.19
  - CVE-2019-12402
  - sonatype-2018-0293
- Upgrade hadoop-common to 2.8.5
  - CVE-2018-11767
- Upgrade hadoop-mapreduce-client-core to 2.8.5
  - CVE-2017-3166
- Upgrade hibernate-validator to 5.2.5
  - CVE-2017-7536
- Upgrade httpclient to 4.5.10
  - sonatype-2017-0359
- Upgrade icu4j to 55.1
  - CVE-2014-8147
- Upgrade jackson-databind to 2.6.7.3:
  - CVE-2017-7525
- Upgrade jetty-http to 9.4.12:
  - CVE-2017-7657
  - CVE-2017-7658
  - CVE-2017-7656
  - CVE-2018-12545
- Upgrade log4j-core to 2.8.2
  - CVE-2017-5645:
- Upgrade netty to 3.10.6
  - CVE-2015-2156
- Upgrade netty-common to 4.1.42
  - CVE-2019-9518
- Upgrade netty-codec-http to 4.1.42
  - CVE-2019-16869
- Upgrade nimbus-jose-jwt to 4.41.1
  - CVE-2017-12972
  - CVE-2017-12974
- Upgrade plexus-utils to 3.0.24
  - CVE-2017-1000487
  - sonatype-2015-0173
  - sonatype-2016-0398
- Upgrade postgresql to 42.2.8
  - CVE-2018-10936

Note that if users are using JDBC lookups with postgres, they may need
to update the JDBC jar used by the lookup extension.

* Fix license for postgresql
2019-11-19 09:14:33 -08:00
Atul Mohan 8515a03c6b Modify batch index task naming to accomodate simultaneous tasks (#8612)
* Change hadoop task naming

* Remove unused

* Add timestamp

* Fix build
2019-11-18 15:07:16 -08:00
Chi Cao Minh d60978343a Improve missing JDBC driver error for lookups (#8872)
If the JDBC drivers are missing from the lookup extensions, throw an
exception that directs the user how to resolve the issue. This change is
a follow up to #8825.
2019-11-18 11:42:38 -08:00
Rye ea8e4066f6 Use earliest offset on kafka newly discovered partitions (#8748)
* Use earliest offset on kafka newly discovered partitions

* resolve conflicts

* remove redundant check cases

* simplified unit tests

* change test case

* rewrite comments

* add regression test

* add junit ignore annotation

* minor modifications

* indent

* override testableKafkaSupervisor and KafkaRecordSupplier to make the test runable

* modified test constructor of kafkaRecordSupplier

* simplify

* delegated constructor
2019-11-18 11:05:31 -08:00
Jihoon Son 1611792855
Add InputSource and InputFormat interfaces (#8823)
* Add InputSource and InputFormat interfaces

* revert orc dependency

* fix dimension exclusions and failing unit tests

* fix tests

* fix test

* fix test

* fix firehose and inputSource for parallel indexing task

* fix tc

* fix tc: remove unused method

* Formattable

* add needsFormat(); renamed to ObjectSource; pass metricsName for reader

* address comments

* fix closing resource

* fix checkstyle

* fix tests

* remove verify from csv

* Revert "remove verify from csv"

This reverts commit 1ea7758489.

* address comments

* fix import order and javadoc

* flatMap

* sampleLine

* Add IntermediateRowParsingReader

* Address comments

* move csv reader test

* remove test for verify

* adjust comments

* Fix InputEntityIteratingReader

* rename source -> entity

* address comments
2019-11-15 09:22:09 -08:00
Jonathan Wei 75ea0d592a Add more datasketches doubles sketch SQL functions (#8843)
* Add more datasketches doubles sketch SQL postaggs

* style and lgtm
2019-11-08 18:05:06 -08:00