Commit Graph

13706 Commits

Author SHA1 Message Date
Simon Willnauer f77804dad3 Bake in TieredMergePolicy
Today we provide the ability to plug in MergePolicy and
we provide the once lucene ships with. We do not recommend to change
the default and even only a small number of expert users would ever touch
this. This commit removes the ancient log byte size and log doc count
merge policy providers, simplifies the MergePolicy wiring and makes the
tiered MP the one and only default. All notions of a merge policy has been
removed from the docs and should be deprecated in the previous version.

Closes #11588
2015-06-11 11:58:30 +02:00
Britta Weber e7a99501b0 [TEST] remove node from nodes list if disruption is removed from node
If we don't remove the node from the nodes list then later clearDisruption might fail
in case we shut down the node before.
2015-06-11 09:08:10 +02:00
Boaz Leskes 134d898be9 Test: added awaitFix to AzureTwoStartedNodesTest.two_nodes_should_run_using_private_ip 2015-06-11 08:59:44 +02:00
Ryan Ernst d372bf7d7a Merge pull request #11586 from rjernst/fix/field-names-null
Mappings: Shortcut exists and missing queries when no types/docs exist
2015-06-10 14:58:57 -07:00
Clinton Gormley 6e71f60b82 Update bool-query.asciidoc
Emphasise section about using bool query in filter context
2015-06-10 21:46:23 +02:00
Simon Willnauer 657d6dd9cf Remove MergeScheduler pluggability
Nobody should really plug in a different merge scheduler for elasticsearch.
This is too expert and might cause catastrophic failures.
2015-06-10 20:28:30 +02:00
Boaz Leskes 650950c7f4 Merge pull request #11583 from bleskes/mapper_exception_unwrap
Recovery: Fix MapperException detection during translog ops replay
2015-06-10 19:06:53 +02:00
Ryan Ernst de4295cd7f Mappings: Shortcut exists and missing queries when no types/docs exist
There used to be a null check for _field_names mapper not existing. This
was recently removed. However, there is a corner case when the mapper
may be missing: when no types or docs exist at all in the index.

This change adds back a null check and just returns no docs.
2015-06-10 09:51:43 -07:00
Igor Motov 831cfa52d5 Snapshot/Restore: change metadata file format 2015-06-10 11:13:32 -04:00
Boaz Leskes e9d275d3a9 Recovery: fix MapperException detection during translog ops replay
The current ExceptionsHelper.unwrapCause(exception) requires the incoming exception to support ElasticsearchWrapperException , which TranslogRecoveryPerformer.BatchOperationException doesn't implement. I opted for a more generic solution
2015-06-10 17:12:14 +02:00
Boaz Leskes 5951f2580d Merge pull request #11576 from bleskes/wait_on_mapping_remove
Test: rename waitForConcreteMappingsOnAll & waitForMappingOnMaster to assertConcreteMappingsOnAll & assertMappingOnMaster
2015-06-10 14:37:59 +02:00
Steve Kearns 39acd7c170 Add Aspire for Elasticsearch to integrations page
Add Aspire for Elasticsearch to the integrations page, and add two missing colons to other integrations.
2015-06-10 08:12:11 -04:00
Adrien Grand 199c9002e5 Merge pull request #11428 from jpountz/fix/CompressedXConten_fast
Make CompressedXContent.equals fast again.
2015-06-10 13:38:06 +02:00
Adrien Grand fbe617c37b Internal: Make CompressedXContent.equals fast again.
We had to make CompressedXContent.equals decompress data to fix some
correctness issues which had the downside of making equals() slow. Now we store
a crc32 alongside compressed data which should help avoid decompress data in
most cases.

Close #11247
2015-06-10 13:37:08 +02:00
Boaz Leskes 356f0ffde2 Test: rename waitForConcreteMappingsOnAll & waitForMappingOnMaster to assertConcreteMappingsOnAll & assertMappingOnMaster
Now that mapping updates are sync and done before indexing we don't really need the waiting component. Also, removed many places were they were used as safe guard against delayed mapping updates, which are now not needed.
2015-06-10 13:27:57 +02:00
Adrien Grand ac7ce2b899 Rivers removal.
While we had initially planned to keep rivers around in 2.0 to ease migration,
keeping support for rivers is challenging as it conflicts with other important
changes that we want to bring to 2.0 like synchronous dynamic mappings updates.
Nothing impossible to fix, but it would increase the complexity of how we
deal with dynamic mappings updates and manage rivers, while handling dynamic
mappings updates correctly is important for resiliency and rivers are on the go.
So removing rivers in 2.0 may well be a better trade-off.
2015-06-10 09:22:09 +02:00
Ryan Ernst 49bef19878 Merge pull request #11565 from rjernst/remove/field-mapper-wrappers
Mappings: Remove leftover sugar methods from FieldMapper
2015-06-09 14:12:08 -07:00
Ryan Ernst 6c73647325 Mappings: Remove leftover sugar methods from FieldMapper
These methods are now all in MappedFieldType. This removes the remaining
callers of the methods on FieldMapper, and cuts down the FieldMapper
API to no longer include them.
2015-06-09 14:02:06 -07:00
Honza Král 09b5f90779 [API] correct link for flush_synced docs 2015-06-09 21:29:02 +02:00
Simon Willnauer 63ed54017e [TEST] Remove deprecated API 2015-06-09 20:43:01 +02:00
Ryan Ernst 9d57818cee fix field names to be disabled again pre 1.3 2015-06-09 11:42:06 -07:00
Simon Willnauer b0466bbacd [TEST] Add simple test for selecting fiels to be included in mlt query 2015-06-09 20:16:47 +02:00
Ryan Ernst 9fa69bd663 Merge pull request #11559 from rjernst/pr/field-type-users
Mappings: Make index level mapping apis use MappedFieldType
2015-06-09 11:04:33 -07:00
Ryan Ernst 1fdae75025 Mappings: Make index level mapping apis use MappedFieldType
The MapperService is the "index wide view" of mappings. Methods on it
are used at query time to lookup how to query a field. This
change reduces the exposed api so that any information returned
is limited to that api exposed by MappedFieldType. In the future,
MappedFieldType will be guaranteed to be the same across all
document types for a given field.

Note CompletionFieldType needed some more settings moved to it. Other
than that, this change is almost purely cosmetic.
2015-06-09 10:04:28 -07:00
Ryan Ernst 9787266b60 Merge pull request #11544 from rjernst/refactor/null-value
Mapping: Move null value handling into MappedFieldType
2015-06-09 09:46:56 -07:00
Ryan Ernst bf805168e9 Merge branch 'master' into refactor/null-value 2015-06-09 09:46:19 -07:00
Ryan Ernst 6719b1b79e add javadocs to null value functions 2015-06-09 09:46:05 -07:00
Robin Clarke f13c216aa2 More information about 'Copy field to' 2015-06-09 16:35:49 +02:00
Alexander Reelsen a54d4e4aa8 Versioning: Adding 1.6.1 development version & 1.6.0 bwc index 2015-06-09 16:30:02 +02:00
Simon Willnauer 54815f259e add lang-javascript module 2015-06-09 15:17:27 +02:00
Simon Willnauer 64fdb0eeb0 migrate branch for lang-javascript 2015-06-09 15:17:27 +02:00
Simon Willnauer a24c10c703 add lang-javascript module 2015-06-09 15:17:25 +02:00
Simon Willnauer d9fe7cc252 add lang-python module 2015-06-09 15:17:23 +02:00
Simon Willnauer a04a9777d4 migrate branch for lang-python 2015-06-09 15:17:23 +02:00
Simon Willnauer a1b410652d add lang-python module 2015-06-09 15:17:22 +02:00
Simon Willnauer ea07e5ba94 Modify script to only migrate plugins 2015-06-09 15:16:42 +02:00
Britta Weber 2edff1bc4a [TEST] stop disruption before we check for deleted shards
The shard can potentially not be deleted if the obsever that checks for the shard
STARTED is not registered because the registering is delayed by the disruption.
If the sum of delays is more than 10s then the wait for shard deletion will time out.
2015-06-09 14:28:46 +02:00
Alexander Reelsen 3bda78e43b ResourceWatcher: Rename settings to prevent watcher clash
The ResourceWatcher used settings prefixed `watcher.`, which
potentially could clash with the watcher plugin.

In order to prevent confusion, the settings have been renamed to
`resource.reload` prefixes.

This also uses the deprecation logging infrastructure introduced
in #11033 to log deprecated settings and their alternative at
startup.

Closes #11175
2015-06-09 10:02:49 +02:00
Alexander Reelsen 57a94a151d IdsQueryBuilder: Allow to add a list in addition to array
In case a developer gets a list of ids from another data source,
it does not make a lot of sense, to convert it to an array first,
and then internally in IdsQueryBuilder elasticsearch creates a
list out of this.

Closes #5089
2015-06-09 09:31:19 +02:00
Chelsea Lura 3ac19e8f7f Doc: Typo 'good' vs 'well' typo
Closes #11549
2015-06-09 09:25:23 +02:00
Michael McCandless 405ab4f78e Merge pull request #11546 from mikemccand/segment_infos_trace
Add node setting to send SegmentInfos debug output to System.out
2015-06-08 18:48:51 -04:00
Michael McCandless 18609fff0d add node setting to send SegmentInfos debug output to System.out 2015-06-08 18:40:58 -04:00
Ryan Ernst ba15a2f6cb Mapping: Move null value handling into MappedFieldType
In order for exists queries to use the null value for
a field, null value needs to be part of the field type (should
differ between document types). This change moves null value
into the field type, as well as simplifies the null value
methods available to remove supportsNullValue().
2015-06-08 15:08:24 -07:00
Boaz Leskes 16d9480d78 Merge pull request #11536 from bleskes/recovery_translog_op_count_on_mapping_retry
Fix recovered translog ops stat counting when retrying a batch
2015-06-08 18:17:50 +02:00
Clinton Gormley 60c7e0eb91 Update merge.asciidoc
Corrected typo in merge docs
2015-06-08 16:45:59 +02:00
Nirmal Chidambaram 931b9f9c74 Filtered out non data-nodes in relevant cat api
Closes #9214
Closes #9287
2015-06-08 16:05:42 +02:00
Colin Goodheart-Smithe 240155b4ea Adds eclipse settings files to set correct encoding and complier preferences
This is done in the maven-eclipse-plugin so now all eclipse files are generated by mV eclipse:eclipse and can safely be ignored
2015-06-08 14:54:44 +01:00
Robert Muir 5241eda8f4 Merge pull request #11534 from rmuir/lucene_5_2
use released lucene 5.2 jar
2015-06-08 09:52:15 -04:00
javanna 1b496d09c3 [TEST] moved custom query parser tests to proper location 2015-06-08 15:50:43 +02:00
Boaz Leskes 10adb71445 Recovery: fix recovered translog ops stat counting when retrying a batch
#11363 introduced a retry logic for the case where we have to wait on a mapping update during the translog replay phase of recovery. The retry throws or recovery stats off as it may count ops twice.
2015-06-08 15:32:06 +02:00