Commit Graph

11056 Commits

Author SHA1 Message Date
javanna 481ef7d3f4 [TEST] add REST test for update index settings api using indices options
Relates to #10030
2015-03-21 11:10:24 +01:00
olivier bourgain 9561678ca0 Update Settings api: fix handling of IndicesOptions parameters in REST layer
The Update Settings API tries to merge the query_string params with the settings sent as body and excluding some "well known params" such as `pretty`, `timeout` etc. Those well known params do not include the params used by IndicesOptions though, so they get merged resulting in invalid settings that get rejected.

Closes #10030
2015-03-21 10:59:44 +01:00
javanna 65b9a94baa Benchmark api: removed leftovers
Closes #10182
Closes #10184
2015-03-21 10:36:05 +01:00
Simon Willnauer 7257345db9 Revert Benchmark API
The benchmark api is being worked on feature/bench branch and will be merged from there when ready.
2015-03-21 10:36:04 +01:00
javanna 66fa72f21c Java API: remove duplicated consistency level setters
`setConsistencyLevel` setter is already present in the base class `ShardReplicationOperationRequestBuilder`. It is not needed in `DeleteRequestBuilder` and `IndexRequestBuilder`.

Closes #10188
2015-03-21 10:30:36 +01:00
Asimov4 649e3aa4c5 [DOCS] Fix typos in percolate.asciidoc 2015-03-21 10:23:15 +01:00
popovae b7328a4d45 [DOCS] Add ElasticOcean mobile app to monitoring.asciidoc 2015-03-21 10:03:56 +01:00
Reuben Sutton 051bc8c49e Rest: remove indices_boost _search URL param
Relates to https://github.com/elastic/elasticsearch-ruby/issues/29, the parsing logic is incorrect as comma is used twice as a separator (e.g. indices_boost=index1,5,index,10 in expected but will never be parsed correctly). Anyways `indices_boost` makes more sense in the request body only where properly parsed and supported.

Closes #6281
2015-03-21 09:25:46 +01:00
javanna 88e506e58c [DOCS] add -i flag to more curl HEAD calls 2015-03-21 08:56:20 +01:00
Corey Daley 366f01b4d2 [DOCS] add -i flag to curl HEAD call
without -i you never see the status:200 or status:404 messages
2015-03-21 08:56:12 +01:00
Robert Muir 6da99b3ef0 [Bootstrap] Throw exception if the JVM will corrupt data.
Detect the worst-offenders, all IBM versions and several known hotspot
versions that can cause index corruption, and fail on startup.

Provide/detect compiler workarounds when they exist, but warn about
performance degradation.

In all cases the check can be bypassed completely with a safety
switch via undocumented system property (es.bypass.vm.check=true)

Closes #7580
2015-03-21 02:47:44 -04:00
Simon Willnauer 5e5f00de08 Read first bytes as ints to respect method contract 2015-03-20 17:09:07 -07:00
Florian Hopf 865cbeb3d8 Filter indices stats for translog
Added the missing call in the RestAction, closes #8262
2015-03-20 14:38:49 -07:00
Shay Banon 31b63b1a84 Better detection of CBOR
CBOR has a special header that is optional, if exists, allows for exact detection. Also, since we know which formats we support in ES, we can support the object major type case.
closes #7640
2015-03-20 22:06:03 +01:00
Al Lefebvre 94f82368f0 Update templates.asciidoc
I've been attempting to programatically verify that adding index templates via the `{path.conf}/templates/` directory works fine although I was never able to validate this via an API call to the `/_template/`.  It seems that these templates do not appear in that API call, which I discovered in the following mail thread:

http://elasticsearch-users.115913.n3.nabble.com/Loading-of-index-settings-template-from-file-in-config-templates-td4024923.html#d1366317284000-912

My question is why wouldn't the `/_template/*` method return these templates?  This tends to complicate things for those that want to perform automated tests to verify that they are in fact being recognized and used by Elasticsearch.
2015-03-20 14:52:20 -06:00
Simon Willnauer 1d357cb48a Merge pull request #10157 from spinscale/1503-cleanup-remove-bytesarray-unsafe
Cleanup: Remove unsafe field in BytesStreamInput
2015-03-20 13:14:02 -07:00
Michael McCandless 34b397597c Core: increase default rate limiting for snapshot, restore and recovery to 40 MB/sec
This also fixes a possible issue that may cause over-throttling when
there are many small files being copied, which should be rare.

Closed #10185
2015-03-20 16:09:42 -04:00
Simon Willnauer 93fedcbb88 [RECOVERY] Wipe shard state before switching recovered files live
Today we leave the shard state behind even if a recovery is half finished
this causes in rare conditions shards to be recovered and promoted as
primaries that have never been fully recovered.

Closes #10053
2015-03-20 08:42:03 -07:00
Boaz Leskes 0f2d2d0495 Test: testCancelRecoveryAndResume - add network hook before bumping replicas 2015-03-20 16:24:36 +01:00
Boaz Leskes 7d64877326 remove left over recovery log 2015-03-20 16:10:26 +01:00
javanna 9ca74c8217 [DOCS] clarify no-master-block docs
Closes #9739
2015-03-20 15:58:16 +01:00
Boaz Leskes 8c69535580 Recovery: add throttle stats
This commit adds throttling statistics to the index stats API and to the recovery state.

Closes #10097
2015-03-20 10:13:40 +01:00
javanna 4348959f9d Delete api: remove broadcast delete if routing is missing when required
This commit changes the behaviour of the delete api when processing a delete request that refers to a type that has routing set to required in the mapping, and the routing is missing in the request. Up until now the delete api sent a broadcast delete request to all of the shards that belong to the index, making sure that the document could be found although the routing value wasn't specified. This was probably not the best choice: if the routing is set to required, an error should be thrown instead.

A `RoutingMissingException` gets now thrown instead, like it happens in the same situation with every other api (index, update, get etc.). Last but not least, this change allows to get rid of a couple of `TransportAction`s, `Request`s and `Response`s and simplify the codebase.

Closes #9123
Closes #10136
2015-03-20 09:19:43 +01:00
Simon Willnauer 780fe57a2c [TEST] Add more logging to TruncatedRecoveryTests 2015-03-19 23:17:48 -07:00
Lee Hinman f14e226e82 [TEST] Add additional logging to testCorruptTranslogFiles 2015-03-19 22:58:49 -06:00
Lee Hinman 25b09ede16 [TEST] Use less shards in testShadowReplicaNaturalRelocation
This makes the assertion a bit more flexible and removes the
`ensureGreen` in favor of `ensureYellow`, which is really all that is
needed to perform a search. On slow machines the relocations can take a
while and time out the `ensureGreen`.
2015-03-19 18:12:21 -06:00
Britta Weber f253542abb remove unused method. close() is neither needed nor called anywhere
closes #10175
2015-03-19 17:05:24 -07:00
Simon Willnauer 48388dbb6e apply feedback 2015-03-19 15:56:28 -07:00
Simon Willnauer d9be606c5e [LIFECYCLE] Add before/afterIndexShardDelete callback
This commit allows code to be executed before or after a shards content
is deleted from disk. This is only executed if the shard owns the
content ie. on a shard file system only a primary shard will execute
these operations.
2015-03-19 15:47:51 -07:00
Simon Willnauer a6897aa073 [GATEWAY] Move GatewayShardsState logic into IndexShard
The index shard should take care of shard state persistence since it has
all the information and the gateway concept has been removed in master.
2015-03-19 14:48:02 -07:00
Simon Willnauer 1168347b9d [REPLICATION] Remove `async` replication
Closes #10114
2015-03-19 14:44:21 -07:00
Boaz Leskes 955ff05a5e Transport: fix potential NPE in new tracer log if request timeout
Closes #9994
2015-03-19 21:29:34 +01:00
Clinton Gormley a8e5c6eeb0 Merge pull request #10169 from clintongormley/deprecate_thrift_memcached
Remove references to the thrift and memcached transport plugins
2015-03-19 20:56:25 +01:00
Clinton Gormley aa94ced0ae Remove references to the thrift and memcached transport plugins
as they are no longer supported

Closes #10166
2015-03-19 20:49:58 +01:00
olivier bourgain deade2eb71 return an HTTP code representing the error when a suggest request failed instead of 200 2015-03-19 13:00:31 -04:00
Clinton Gormley 6e9eea9b98 Deleted REST test pending shard header
Bug introduced in #10162
2015-03-19 17:57:25 +01:00
Clinton Gormley ef5bf9c21b Changed the release notes script to use the category labels instead of title prefix 2015-03-19 16:52:47 +01:00
Clinton Gormley 806e7b4763 Merge pull request #10162 from clintongormley/deprecate_async_master
Remove async replication from the docs and REST spec
2015-03-19 15:35:23 +01:00
Clinton Gormley 25369f0727 Remove async replication from the docs and REST spec
Relates to #10114
2015-03-19 15:34:12 +01:00
javanna f4691458d5 [DOCS] added note about dynamic scriptings and updated links in getting started page
Closes #10074
2015-03-19 15:22:39 +01:00
Alexander Reelsen 16efc50139 Cleanup: Remove unsafe field in BytesStreamInput
It is only used once (otherwise it is set to `false` anyway), so copying the array in that case makes more sense.
2015-03-19 13:05:59 +01:00
javanna ddcecc2bc2 [DOCS] added instructions on how to write parameterized tests
Closes #9423
2015-03-19 12:43:51 +01:00
Simon Willnauer 5d39bea91b [TEST] Add tests that triggers truncated files on recovery
Several issues where reported showing truncated files where footers
didn't match and checksums read past EOF. This test reproduces the issue
on the latest 1.4 branch but passes on all versions above.

Closes #10155
2015-03-19 01:03:47 -07:00
Nicholas Knize 9bd0c019f2 [TEST] Fix geo orientation integration test
Pass node name to getInstance method of internalCluster to ensure we return the indexService from a node containing the index.
2015-03-18 21:30:20 -05:00
Nicholas Knize 1a79e1c8cc [GEO] Fix orientation persistence
Fixing geo_shape field mapper to persist the orientation parameter. Also adding parsing and integration tests to ensure persistence across cluster restarts.
2015-03-18 18:55:49 -05:00
Christoph Büscher 5788289a6d [TEST] Remove 'gtelte' feature from yaml tests
Some yaml tests still used the 'gtelte' feature in their skip section.
Since all language clients support this, the feature skip can be
removed.
2015-03-18 15:31:17 -07:00
jaymode 105bdd486a [HTTP] add option to only return simple exception messages
Adds a setting to disable detailed error messages and full exception stack traces
in HTTP responses. When set to false, the error_trace request parameter will result
in a HTTP 400 response. When the error_trace parameter is not present, the message
of the first ElasticsearchException will be output and no nested exception messages
will be output.
2015-03-18 17:49:05 -04:00
Robert Muir 5f1ed47df6 Merge pull request #10143 from rmuir/bwcompat1
use bigger back compat indexes
2015-03-18 14:49:32 -04:00
Robert Muir 3a0700862a regenerate 0.90.x indexes without completion suggester 2015-03-18 14:29:41 -04:00
Robert Muir 087b107dd2 Create bigger back compat indexes. 2015-03-18 14:16:16 -04:00