Commit Graph

2490 Commits

Author SHA1 Message Date
Martijn van Groningen 467a47670c Merge remote-tracking branch 'es/master' into feature/ingest 2015-11-30 10:24:27 +01:00
Robert Muir b7523bb974 Add workaround for JDK-6427854
See e.g. http://build-us-00.elastic.co/job/es_feature_ingest/2831/consoleFull

The bug can still happen, so we should let netty do this workaround
2015-11-29 16:12:52 -05:00
Michael McCandless 94954ece3a improve javadocs 2015-11-29 06:13:47 -05:00
Michael McCandless 8933947b95 only pull Fields once from the reader 2015-11-29 06:00:11 -05:00
javanna b4b698f653 Merge branch 'master' into feature/ingest 2015-11-28 11:15:25 +01:00
Robert Muir 29853744a3 Merge pull request #15055 from rmuir/fun_with_windows
set ActiveProcessLimit=1 on windows
2015-11-27 16:57:34 -05:00
Christoph Büscher b375084305 Tests: Correction in AbstractQueryTestCase
Removed check that two query builder that are different according
to equals() have different hashCode since that is not required
by the contract of hashCode.
2015-11-27 17:36:57 +01:00
Daniel Mitterdorfer 2f87e0a4d9 Declare NetworkService non-final to support mocking in tests 2015-11-27 15:51:19 +01:00
Robert Muir ee827198ff set ActiveProcessLimit=1 on windows
This blocks process creation, similar to the approach for BSD.
2015-11-27 01:50:09 -05:00
Robert Muir aa56bfa562 Merge pull request #15039 from rmuir/setrlimit
set RLIMIT_NPROC = 0 on bsd/os X systems.
2015-11-26 21:18:57 -05:00
Jason Tedor 5c5e02e0bf Remove outdated comment and cleanup RestVerifyRepositoryAction 2015-11-26 19:11:11 -05:00
Simon Willnauer e363e0c2aa Merge pull request #15012 from s1monw/issues/14866
Prevent writing to closed channel if translog is already closed
2015-11-26 14:47:38 +01:00
Martijn van Groningen afc9069c99 * Inlined PipelineStoreClient class into the PipelineStore class
* Moved PipelineReference to a top level class and named it PipelineDefinition
* Pulled some logic from the crud transport classes to the PipelineStore
* Use IOUtils#close(...) where appropriate
2015-11-26 14:42:39 +01:00
Simon Willnauer 59be738e40 Merge pull request #15035 from s1monw/simplify_monitoring_service
Simplify MonitorService construction and detach from guice
2015-11-26 14:05:32 +01:00
Simon Willnauer 9f6598b18d Fix compile errors 2015-11-26 13:41:00 +01:00
Britta Weber 16b84b3999 run bwc test also as integ test and share methods
We had no itegration test before with long terms and several shards only
a bwc test.

related to #14948
2015-11-26 10:57:33 +01:00
Simon Willnauer 3dfa146632 Move ensureOpen calls under lock
We used to check on several places if we are still open but non of these
places did the check under the lock which leaves a small window where we
potentially get closed but still access an already closed channel or another
IO resource.
2015-11-26 10:14:45 +01:00
Simon Willnauer dfdc738a9f Simplify MonitorService construction and detach from guice
This is a pretty trivial change that moves most of the monitor service related
object creation from guice into the monitor service. This is a babystep towards removing
guice on the node level as well. Instead of opening huge PRs I try to do this in baby-steps
that are easier to digest.
2015-11-26 09:50:41 +01:00
Robert Muir 2341330aba set RLIMIT_NPROC = 0 on bsd/os X systems.
This BSD-specific limit prevents child process creation.
2015-11-26 02:58:53 -05:00
Christoph Büscher 609d9db470 Geo: Fix toString() in GeoDistanceRangeQuery and GeoPolygonQuery
Minor typo in the two queries toString() method. They were former
filters the toString() seems to have been forgotten while renaming.
2015-11-25 21:52:20 +01:00
Simon Willnauer 9c1f930b8c Prevent writing to closed channel if translog is already closed
We handle AlreadyClosedExceptions gracefully wherever IndexShard / Engine
is used. In some cases, instead of throwing the appropriate exception we
bubble up ChannelClosedException instead which causes shard failures etc.
Today, it seems like that this can only happen if the engine is closed without
acquireing the lock which means that the shard has failed already so the impact is really
just a confusing log message. Yet, this change enforces throwing the right exception
if the translog is already closed.

Closes #14866
2015-11-25 13:54:13 +01:00
Adrien Grand e2447c7ce0 Merge pull request #14981 from jpountz/fix/assert_version_sanity
Check that the declared versions match the backward indices.
2015-11-25 10:49:30 +01:00
Isabel Drost-Fromm 473b19400f Merge pull request #14827 from MaineC/bug-fix/10021-error-listeners
Adds exception objects to log messages.
2015-11-25 10:42:38 +01:00
Christoph Büscher e507ef807a Merge pull request #14933 from cbuescher/shapes-refactor-point
Make PointBuilder, CircleBuilder & EnvelopeBuilder implement Writable
2015-11-25 09:57:20 +01:00
Christoph Büscher 1f94448e92 Making EnvelopeBuilder writable and adding equals/hashCode 2015-11-25 08:43:25 +01:00
Adrien Grand 5fe3ce89e0 Check that the declared versions match the backward indices. 2015-11-24 20:40:02 +01:00
Clinton Gormley c512cc5248 Added bwc indices for 2.0.1 and 2.1.0 and versions 2.0.2-SNAPSHOT and 2.1.1-SNAPSHOT 2015-11-24 19:32:18 +01:00
Yannick Welsch ff9c33b6c8 Use ObjectParser to parse AllocationID 2015-11-24 12:13:54 +01:00
Britta Weber 556d80501c aggs: fix significant terms reduce for long terms
Significant terms were not reduced correctly if they were long terms.
Also, clean up the bwc test a little. Upgrades are not needed.

related to #13522
2015-11-24 11:04:48 +01:00
Adrien Grand 5f33fbdb75 Register field mappers at the node level.
This moves the registration of field mappers from the index level to the node
level and also ensures that mappers coming from plugins are treated no
differently from core mappers.
2015-11-24 08:59:37 +01:00
Jason Tedor b31f2850c0 Do not be lenient when parsing CIDRs
This commit fixes some leniency in the parsing of CIDRs. The leniency
that existed before includes not validating the octet values nor
validating the network mask; in some cases issues with these values were
silently ignored. Parsing is now done according to the guidelines in RFC
4632, page 6.

Closes #14862
2015-11-23 20:46:23 -05:00
Michael McCandless e13b0d4bde upgrade lucene to 5.4.0-snapshot-1715952 2015-11-23 17:13:49 -05:00
Jason Tedor dcfdc0dbf3 Remove leftover TODOs in o.e.c.r.RoutingTable 2015-11-23 12:57:06 -05:00
Yannick Welsch 3dcd35d1c8 Fix issue where shard state metadata is written while containing no allocation id
Such a write can happen when upgrading shard state metadata using the MultiDataPathUpgrader

Relates to #14831
2015-11-23 18:14:02 +01:00
Daniel Mitterdorfer b804a0c4a6 Improve stability of UpdateIT
With this commit, we reduce the amount of work
that UpdateIT does and add progress logging.

Closes #14877
2015-11-23 17:54:57 +01:00
Yannick Welsch fccad13029 Persist allocation ID with shard state metadata on nodes
Closes #14831
2015-11-23 17:35:33 +01:00
Isabel Drost-Fromm 37d7629146 Merge pull request #14249 from MaineC/enhancement/8964
Switch query parsers to use ParseField
2015-11-23 14:16:40 +01:00
Boaz Leskes c1264b926f Improve trace logging in TransportReplicationAction and error reporting at RecoveryWhileUnderLoadIT
Things that helped me traced down an issue.

Closes #14931
2015-11-23 13:52:32 +01:00
Boaz Leskes 4a6f3c7840 Make sure the remaining delay of unassigned shard is updated with every reroute
For example: if a node left the cluster and an async store fetch was triggered. In that time no shard is marked as delayed (and strictly speaking it's not yet delayed). This caused test for shard delays post node left to fail. see : http://build-us-00.elastic.co/job/es_core_master_windows-2012-r2/2074/testReport/

 To fix this, the delay update is now done by the Allocation Service, based of a fixed time stamp that is determined at the beginning of the reroute.

 Also, this commit fixes a bug where unassigned info instances were reused across shard routings, causing calculated delays to be leaked.

Closes #14890
2015-11-23 13:16:15 +01:00
Isabel Drost-Fromm 38e1ad5596 Switches from manual field parsing to ParseField
This switches query parsing from manual field parsing to using ParseField.

Also adds unit tests for each query that check original json can be parsed
into query builders.

Relates to #8964
2015-11-23 11:37:46 +01:00
Christoph Büscher e25f7042b9 Making CircleBuilder writable and adding equals/hashcode 2015-11-23 11:35:44 +01:00
Christoph Büscher 5ca9f3ff8f Geo: Make ShapeBuilders implement Writable
We recently refactored the queries to make them parsable on the
coordinating note and adding serialization and equals/hashCode
capability to them. So far ShapeBuilders nested inside queries
were still transported as a byte array that needs to be parsed
later on the shard receiving the query. To be able to also
serialize geo shapes this way, we also need to make all the
implementations of ShapeBuilder implement Writable.

This PR adds this to PointBuilder and also adds tests for
serialization, equality and hashCode.
2015-11-23 11:24:43 +01:00
Boaz Leskes 6e2e91cf2a Set an newly created IndexShard's ShardRouting before exposing it to operations
The work for #10708 requires tighter integration with the current shard routing of a shard. As such, we need to make sure it is set before the IndexService exposes the shard to external operations.

Closes #14918
2015-11-23 11:22:40 +01:00
Dominik Stadler 72cab4e9b6 Map log-level 'trace' to JDK-Level 'FINEST' to make it possible to actually enable the trace-log via JdkESLogger.setLevel() 2015-11-23 09:49:29 +01:00
Martijn van Groningen 48771f1a76 field stats: Added `min_value_as_string` and `max_value_as_string` response elements for all number based fields. The existing `min_value` and `max_value` will return the values as numbers instead.
Closes #14404
2015-11-23 08:48:28 +01:00
Martijn van Groningen 4b94b50eb1 field stats: Index constraints should remove indices in the response if the field to evaluate s empty
Index constraints should remove indices in the response if the field to evaluate if empty. Index constraints can't work with that and it is the same as if the field doesn't match.
2015-11-23 08:35:15 +01:00
Jason Tedor 5e2eb7d0de Clarify three possible outcomes in CacheTests#testCachePollution 2015-11-22 19:58:04 -05:00
Jason Tedor 14a635fe4b Use correct equality in CacheTests#testReplaceRecomputesSize.Value 2015-11-22 19:56:27 -05:00
Ryan Ernst b8e462b934 Build: Change project attachment into special extra-plugins dir
Currently we use the "gradle project attachment plugin" to support
building elasticsearch as part of another project. However, this plugin
has a number of issues, a large part of which is requiring consistent
use of the projectsPrefix.

This change removes projectsPrefix, and adds support for a special
extra-plugins directory in the root of elasticsearch. Any projects
checked out within this directory will be automatically added to
elasticsearch.
2015-11-22 08:44:33 -08:00
Robert Muir 5337686d8d Merge pull request #14914 from rmuir/lock_down_system_property_writes
Ban write access to system properties
2015-11-22 11:28:55 -05:00