honeymoose
/
OpenSearch
mirror of https://github.com/honeymoose/OpenSearch.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
1,688 Commits 7 Branches 6 Tags 546 MiB
Commit Graph

1438 Commits

Author SHA1 Message Date
kimchy a3852766ff only check for prefered allocation on data nodes 2010-08-22 00:32:43 +03:00
kimchy 3f701365b0 move the allocate replica after primary active logic to a node allocation 2010-08-22 00:14:56 +03:00
kimchy 5ded04c5cb refactor shard allocation to be more pluggable for node allocation "deciders" 2010-08-21 21:59:31 +03:00
kimchy 2865ceef85 more on sorted nodes when allocating shards above average number of shards per node 2010-08-21 19:55:38 +03:00
kimchy 76e1a6b1bf check also for negative values 2010-08-21 19:51:14 +03:00
kimchy 5d2c005622 pass to canAllocate (though nothing is done there now) the actual routing nodes 2010-08-21 17:52:24 +03:00
kimchy 1ddd969fa7 pass to canAllocate (though nothing is done there now) the actual routing nodes 2010-08-21 17:52:02 +03:00
kimchy 11dbbebb7a renamve var to better reflect its state (started, not active) 2010-08-21 17:50:08 +03:00
kimchy 3117341f44 Shards Allocation: Only rebalance a shard if all its instances are already active, closes #331. 2010-08-21 17:49:19 +03:00
kimchy 38e6649a7e Cluster Health API: Add `initializing_shards` and `unassigned_shards` to the response, closes #330. 2010-08-21 16:03:29 +03:00
kimchy 113ea1bb1b don't throttle gateway recovery based on peer recoveries (gateway recoveries are important to do as fast as possible). still throttle peer recoveries based on both. 2010-08-21 15:37:12 +03:00
kimchy e12bdd9faf rename RETRY stage in recovery to THROTTLE 2010-08-21 15:13:34 +03:00
kimchy 0c72bb2125 better writing of local md5 caches 2010-08-21 02:16:36 +03:00
kimchy 1ef1e9b25b don't spawn on received cluster state 2010-08-20 19:36:39 +03:00
kimchy 800e450e3a handle local md5 cases, don't include them in the list of files the directory has 2010-08-20 19:30:26 +03:00
kimchy 9acae60dbf also have a minimum value for concurrent recoveries, at 3 2010-08-20 17:28:31 +03:00
kimchy f62f7b8ffe check the progress in case the index size has not been set yet... 2010-08-20 17:17:32 +03:00
kimchy 9f0d5f8ef1 better default concurrent recoveries, based on number of processors, tap it at 10 2010-08-20 17:06:56 +03:00
kimchy 8c25be6dee better exception message with primary not started exception on index/delete/delete_by_query apis 2010-08-20 16:36:25 +03:00
kimchy 07285d81d1 add stream done when there is no file (should not happen) 2010-08-20 15:59:46 +03:00
kimchy b609162be3 don't ignore recovery on throttling unless the shard is closed 2010-08-20 15:58:24 +03:00
kimchy 4e74001bde move marking recovering before creating a status 2010-08-20 15:24:49 +03:00
kimchy c0552bdc70 Can't set "no stopwords" on analyzer, closes #329. 2010-08-20 03:22:43 +03:00
kimchy 6f1a9fbfb9 create md5 files on the local index storage as well, so they won't have to be recomputed when performing initial recovery 2010-08-20 02:28:14 +03:00
kimchy acae0650c8 bail out of recovery when close and still waiting on throttling 2010-08-20 01:24:02 +03:00
kimchy a05c367c5e don't snapshot gateway when cleaning shard 2010-08-20 00:59:37 +03:00
kimchy 4a2e076c6a do a reuse work prefetching not on the cluster state thread, so it won't block while fetching md5s from other nodes 2010-08-20 00:45:17 +03:00
kimchy a5c5f65da5 fix recovery status when all index got reused 2010-08-19 22:46:58 +03:00
kimchy 7437acfcea add progress to indices status API 2010-08-19 22:29:31 +03:00
kimchy 8d0e5b239c recator source rest building into common code 2010-08-19 20:24:12 +03:00
kimchy 7b1093fb16 anoter compressed string test 2010-08-19 18:45:15 +03:00
kimchy 19abe7a2a5 when using compressed source, extract the compressed source directly into the response without any buffering 2010-08-19 18:29:36 +03:00
kimchy 46ccee8f89 when deleting an index, remove any blocks associated with it 2010-08-19 17:21:14 +03:00
kimchy c40935ae14 store the mapping definition in compressed form internally to save memory 2010-08-19 17:06:36 +03:00
kimchy a2d10d490d allocate list when deserializing, NPE galore... 2010-08-19 16:40:51 +03:00
kimchy 04ae1581a0 fix NPE introduced on optimization of index status on docs status creation 2010-08-19 15:37:06 +03:00
kimchy 70c344b079 add compressed string 2010-08-19 15:33:57 +03:00
kimchy 1a18387fab add `standard_html_strip` analyzer that combines the standard analyzer with html_strip char filter. 2010-08-19 15:10:31 +03:00
kimchy 22ea5e6608 fix primary election on replica relocation 2010-08-19 14:59:50 +03:00
kimchy 9c0bbe9bb5 log failure to notify master on shard started/failed 2010-08-19 13:49:48 +03:00
kimchy fd569a9523 renme tests to reflect replica and not backup 2010-08-19 13:11:53 +03:00
kimchy 27e92cd505 fix a case where node might not exists on relocating, don't fail the replication action (index / delete / ...) because of it 2010-08-19 13:08:20 +03:00
kimchy 51656552a5 better mehtod name 2010-08-19 11:20:06 +03:00
kimchy 965d7303cf index status action - some docs and extract common docs status 2010-08-19 09:34:02 +03:00
kimchy a3107bc5b1 add snapshot status (either current one or the latest one executed) to indices status api 2010-08-18 17:11:29 +03:00
kimchy 91aada2706 boolean values cannot be searched with expected queries, closes #327. 2010-08-18 15:11:23 +03:00
kimchy de71a1ce99 filtered more_like_this doesn't parse correctly, closes #326. 2010-08-18 14:36:34 +03:00
kimchy 2259ef671b Gateway: Failure to read full translog from the gateway, closes #328. 2010-08-18 14:29:46 +03:00
kimchy 3f9034b41c add gateway recovery status to the indices status API exposing both on going and summary when recovering from a gateway 2010-08-17 22:34:56 +03:00
kimchy 311520d146 add peer recovery status to the indices status API exposing both on going and summary when recovering from a peer shard 2010-08-17 21:33:11 +03:00
kimchy 5fb80c391b add peer recovery status to the indices status API exposing both on going and summary when recovering from a peer shard 2010-08-17 21:23:05 +03:00
kimchy 96fc16dddb add ability to iterate over assigned shards, and add an extension point to control shard routing iteration in the broadcast based action support 2010-08-17 14:04:12 +03:00
kimchy 7833cb1c76 Document IDs need to be URL decoded when indexed, closes #324. 2010-08-17 10:47:02 +03:00
kimchy c989d3a928 fix limit on UTF writing (the new mechanism) to be higher than short 2010-08-17 10:21:26 +03:00
kimchy e1f3fee4e4 internally store on going state of snapshot to gateway 2010-08-17 10:03:19 +03:00
kimchy 29e981d28d internally store on going state of recovery from gateway 2010-08-17 08:17:29 +03:00
kimchy 92fd9af2b9 more info stored on shard recovery process, start time and recovery time 2010-08-16 18:22:19 +03:00
kimchy 92aa91d7b2 update the current recvoerd files size in peer recovery 2010-08-16 09:11:51 +03:00
kimchy a012e0f1bd allow to configure translog batch size (internal) and compress translog operations when doing peer recvoery 2010-08-16 09:03:48 +03:00
kimchy 1bdce4c7ef capture current state of peer level recovery 2010-08-16 09:00:34 +03:00
kimchy 57ee1bdc55 improve utf based stream input / output handling 2010-08-15 21:19:36 +03:00
kimchy 14237317fc support compression of the source field 2010-08-15 17:57:14 +03:00
kimchy ee33ee457a allow to explicitly disable compression in peer recovery (defaults to true) 2010-08-15 15:13:45 +03:00
kimchy f6c58ff35d change default calibrate_size_by_deletes for merge policy to true from false (this is what upcoming lucene version will use, and makes much more sense) 2010-08-15 09:01:09 +03:00
kimchy 6d509a5e4e explicitly clean stream handles 2010-08-15 03:08:01 +03:00
kimchy c18904eb96 Recovery: Using transport compression explicitly when recovering an index from a peer shard, closes #322. 2010-08-15 03:01:35 +03:00
kimchy 1ee2f80e68 Transport: add global compression support compressing all internal transport communication (using lzf), closes #321. 2010-08-15 02:57:22 +03:00
kimchy c10544479f refactor cached streams into a separate class 2010-08-14 13:47:34 +03:00
kimchy e1226626bc improve thread cached bytes allocation 2010-08-14 02:26:12 +03:00
kimchy 7c5f0fe405 minor transport refactoring, add options to send response, though no options to set yet 2010-08-14 02:25:58 +03:00
kimchy b3afca2589 remove compressor package, move lzf to a different package 2010-08-14 02:25:30 +03:00
kimchy 0adb2edba2 clean code 2010-08-14 01:23:04 +03:00
kimchy 75ee01a6b4 clean builder code when dealing with maps 2010-08-14 00:37:33 +03:00
kimchy 4a9c7d672e clean code 2010-08-13 23:50:16 +03:00
kimchy d12c757824 Gateway: Chunk based storage broken, fails to recover from gateway, closes #318. 2010-08-13 14:03:52 +03:00
kimchy 12ef12f7aa better handling of closing a shard / index while retrying recovery 2010-08-12 21:44:27 +03:00
kimchy 2bd9a63467 Query DSL: custom score script, allow to use `_score` as well as `score` as the underlying query score, closes #316. 2010-08-12 18:41:29 +03:00
kimchy 98bc8285ea Analysis: Add `char_filter` on top of `tokenizer`, `filter`, and `analyzer`. Add an `html_strip` char filter, closes #315. 2010-08-12 18:16:30 +03:00
kimchy e29925684a Term Facets: Add `reverse_cout` and `reverse_term` to `order` options, closes #314. 2010-08-12 15:26:26 +03:00
kimchy b1d1f1ff94 improve peer recovery shard state handling. when throttling a recovery, don't restore the non recovering state and then move to recovery again with each retry, keep the recovering state while retrying 2010-08-12 12:31:27 +03:00
kimchy c0a7dc327c range facet: from is used in to_str, should be to 2010-08-11 23:40:42 +03:00
kimchy a6d84f5942 no need to spawn a thread when response to a ping response 2010-08-11 13:20:13 +03:00
kimchy fbea92e688 REST API hang on a bogus call, closes #297. 2010-08-11 13:18:33 +03:00
kimchy ee26d55296 Remove memory monitor and move translog operations threshold to shard level setting, closes #312. 2010-08-11 12:54:00 +03:00
kimchy df4ece8cef remove no longer required commented code ;) 2010-08-11 11:43:52 +03:00
kimchy 7977edd5db a filter based on the type of the document is built in several places in the code, which is wasteful. Instead, create a filter based on the type on the document mapper level, and reuse it where applicable. 2010-08-10 18:34:49 +03:00
kimchy 5f986ef422 build a match all filter internal implementation and use it where appropiate (instead of using query filter wrapper around a match all query) 2010-08-10 18:18:51 +03:00
kimchy f2018e2f86 Facets: Automatically filter by type for facets that use explicit field names that are prefixed by the type, closes #310. 2010-08-10 17:03:14 +03:00
kimchy 61bb9d0ff9 Mappers: also use `name` as a smart lookup for field/property on top of indexName and fullName, closes #309. 2010-08-10 16:14:15 +03:00
kimchy 879191c435 don't add the same object twice under rare cases where multiple threads are introducing the same new dynamic object 2010-08-10 09:18:53 +03:00
kimchy 4bb1baa9d4 remove nb structures 2010-08-10 00:05:38 +03:00
kimchy 05d07b036a remove ning async http client, no need for it 2010-08-10 00:00:24 +03:00
kimchy f3cc8d1249 when serializing mappers into json, also serialize index, type and id mappings (index for the enabled flag, and id/type for the store) 2010-08-09 18:25:45 +03:00
kimchy f10699d5b8 javadoc 2010-08-09 09:06:14 +03:00
kimchy 6611e7db44 refactor custom sort parsing to be "code" pluggable 2010-08-09 08:57:56 +03:00
kimchy ff8d743337 add geo distance sort option, refactor java level sort API to have sort builders (similar to facets) 2010-08-08 21:46:11 +03:00
kimchy 8b19d353c1 add sort values as part of the response per search hit 2010-08-08 18:21:34 +03:00
kimchy 37af7f5aef add sort values as part of the response per search hit 2010-08-08 18:21:17 +03:00
kimchy 7cb2010dda add script support for terms facets, allowing to control if a certain term will be included in the facets result (boolean result), or muging of the term 2010-08-08 10:37:14 +03:00
kimchy ee1d50f8d8 add script support for terms facets, allowing to control if a certain term will be included in the facets result (boolean result), or muging of the term 2010-08-08 10:27:06 +03:00
kimchy 566ae94478 add the ability to control the "order" by of terms facet, can either be `term` or `count`. defaults to `count`. 2010-08-08 09:31:40 +03:00
kimchy d4547c629f New API, update settings, allowing to change specific runtime settings of the cluster (indices). `number_of_replicas` is the only settings supported, allowing to change the number of repliacs of one or more indices. 2010-08-08 09:17:09 +03:00
kimchy 6d3d9fd807 start work on being able to change the number of replicas at runtime 2010-08-05 11:00:02 +03:00
kimchy 860d8058b6 rename jmx settings to conform with casing (old casing also works) 2010-08-04 22:27:26 +03:00
kimchy 8c5fdf4aee Ability to return all stored fields with a *, closes #296. 2010-08-04 13:22:55 +03:00
kimchy d72de60b6f Add the ability to store the index name in the doc, closes #292. 2010-08-04 11:41:56 +03:00
kimchy a44d30bb61 After gateway recovery, mappings keep being applied on each cluster change, closes #295. 2010-08-04 09:02:10 +03:00
kimchy 959eb0e703 Geo: Polygon based filter, closes #294. 2010-08-03 21:52:13 +03:00
kimchy 39f344c11e change the format of bounding box filter to conform more to the "field_name" : { ... } structure 2010-08-03 19:18:16 +03:00
kimchy 91c256361e support for custom script based sorting 2010-08-03 18:14:02 +03:00
kimchy 3d31c38f11 refactoring transport service to allow for send options 2010-08-03 10:06:30 +03:00
kimchy 04781e038a when timeouting, and closed, raise onClose and not onTimeout 2010-08-03 10:05:19 +03:00
kimchy de8ffaf5fb Geo: bounding box filter, closes #290. 2010-08-03 01:51:46 +03:00
kimchy d1e404c1b3 change default geo point resolution from 32 to 64 2010-08-03 00:08:56 +03:00
kimchy 07ab5dcf9b reuse constants for facet type 2010-08-02 22:27:03 +03:00
kimchy dc6ef326d9 Facets: Filter based facet, closes #289. 2010-08-02 19:17:57 +03:00
kimchy 4e6087ee2d improve concurrency when loading field data 2010-08-02 18:20:49 +03:00
kimchy 7719dcd790 add an option to provide the array of ranges "on" the field name itself 2010-08-02 18:09:26 +03:00
kimchy 99af6df0ea add an option to provide the array of ranges "on" the field name itself 2010-08-02 16:39:28 +03:00
kimchy 6ed89ad406 fix comment 2010-08-02 16:35:08 +03:00
kimchy 32e26a6bd0 fix valueField name 2010-08-02 16:32:52 +03:00
kimchy 76cfbe6a44 support dates in range facet 2010-08-02 10:35:21 +03:00
kimchy 4bbbf186b6 fix some javadocs 2010-08-02 09:53:08 +03:00
kimchy 38fe6860d8 update docs 2010-08-02 09:39:17 +03:00
kimchy ef861a6b7b Search Facets: Range Facet, closes #287. 2010-08-01 20:08:27 +03:00
kimchy 9f441d8b9f add filter and global support 2010-08-01 17:09:40 +03:00
kimchy e77d1d0147 add filter and global support 2010-08-01 17:08:44 +03:00
kimchy c5811eb060 add a check for at least one range 2010-08-01 17:05:06 +03:00
kimchy c08791f816 fix type on xcontent result 2010-08-01 16:34:00 +03:00
kimchy 168363106b fix doc 2010-08-01 16:31:54 +03:00
kimchy b8b21a3363 Geo: `geo_distance` facet, closes #286. 2010-08-01 13:31:03 +03:00
kimchy dfb68c6310 Search fails when sorting on a field that has no values in the hits returned, closes #285. 2010-07-31 16:41:15 +03:00
kimchy a26b4f31e1 print the exception when failing to add a specific mapping during cluster state update 2010-07-31 15:41:11 +03:00
kimchy 484665af80 handle cases when there is no value 2010-07-31 02:30:52 +03:00
kimchy 9af288a42f fix test for existance of directory based mappings when creating an index 2010-07-29 19:22:21 +03:00
kimchy 0186a03684 add another script function test 2010-07-29 16:07:57 +03:00
kimchy 4f407e18aa Put Mapping: When using a single node and updating a mapping, it is not marked as `acknowledged`, closes #280. 2010-07-29 15:21:11 +03:00
kimchy e955e41a91 improve handling of failure when trying to read md5, should be ignored on any exception, not just io exception 2010-07-29 15:10:37 +03:00
kimchy 194e6cbff6 Geo support: geo_point type and geo_distance filter, closes #278, #279 2010-07-28 21:06:18 +03:00
kimchy af055f75ce should be static 2010-07-26 13:46:01 +03:00
kimchy f251e2fd32 support lt(N), le(N), gt(N), ge(N) in wait for nodes (thanks Lukas) 2010-07-26 13:44:29 +03:00
kimchy 65284ba2ba Terms Factes: Allow to provide regex controlling which terms should be included, closes #277. 2010-07-26 12:23:03 +03:00
kimchy 03cd2f3772 fix <= 2010-07-26 00:22:21 +03:00
kimchy ac7c8cb650 Analysis: Add pattern analyzer, closes #276. 2010-07-25 22:40:50 +03:00
kimchy 477a24efc6 Mapping: Revise dynamic mapping (into default), merge default to new mappings, closes #275. 2010-07-25 21:31:16 +03:00
kimchy 1884c4219a Mapping: Dynamic mapping definitions are ignored, closes #274. 2010-07-25 12:03:04 +03:00
kimchy 457b56937e small improvements to groovy client, allows for closure in the search request 2010-07-25 10:32:22 +03:00
kimchy 77b6d1d8b8 remove cloud plugin, reimplement an AWS specific cloud plugin with S3 gateway and EC2 discovery 2010-07-24 23:39:43 +03:00
kimchy a6a993ff08 fix ignored exception 2010-07-24 18:48:28 +03:00
kimchy 0072dd816b nicer logging, and use sync streams 2010-07-23 23:10:59 +03:00
kimchy 23ddf8bc53 do a reroute when a node becomes master 2010-07-23 20:55:26 +03:00
kimchy f6255f8ad9 change logging level to debug 2010-07-23 19:29:14 +03:00
kimchy 72682a6730 always retry on connect exception with master operations 2010-07-23 01:59:56 +03:00
kimchy 9283e2a7ad local discovery should update the local metadata state once it started 2010-07-23 01:39:37 +03:00
kimchy 5706e5f6b9 rename node close to node closed exception 2010-07-23 01:12:36 +03:00
kimchy 34b84b0e61 when retrying on primary shard when node disconnected, add it again the the cluster service as a listener 2010-07-23 01:07:09 +03:00
kimchy 2a3130c649 REST Search API: Change `score` to `_score` to denote sorting by hit score, closes #271. 2010-07-22 23:39:55 +03:00
kimchy 42c2b6b743 throw a proper exception when no type is found 2010-07-22 21:17:11 +03:00
kimchy c1ab9f290f remove compile warnings 2010-07-22 00:25:25 +03:00
kimchy b55fd0c0fc verify that data that is written to the blob (in fs) is the same as expected in the blob write api call 2010-07-21 23:46:11 +03:00
kimchy 4dcf219086 after raising a failure, return, don't try and write the blob 2010-07-21 23:45:15 +03:00
kimchy 2432a7cb5b fixed highlighted styled built in schema tags to properly increate from hlt1 to hlt10 2010-07-21 20:38:13 +03:00
kimchy 8ec7ee66a9 Search: Sending a request that fails to parse can cause file leaks, closes #270. 2010-07-21 16:59:58 +03:00
kimchy 0cb97e4044 cancel the timeout in any case 2010-07-21 16:38:37 +03:00
kimchy 8d45901a8a change logging level, nodes FD will detect it as well 2010-07-21 16:30:40 +03:00
kimchy 48d33ec70a Cluster Health API: Add `wait_for_nodes` (accepts "N", "<N", ">N", "<=N", and ">=N"), closes #269. 2010-07-21 16:29:44 +03:00
kimchy 2a0e8d4ec9 better state when closing shard, and handling its state 2010-07-21 09:20:28 +03:00
kimchy 3629540953 when trying to rejoin the cluster, clean the routing/metadata so the node will get fresh shards 2010-07-21 08:55:35 +03:00
kimchy 15cf7a2361 initialize the nodes before a possible retry, so we have the local node id for it 2010-07-21 01:53:00 +03:00
kimchy 39e546b64d add a proper master not discovered exception for master based operations (cluster state) and the ability to control the timeout till master is found 2010-07-21 01:51:58 +03:00
kimchy a2ceaa91cc improve retry when closing the node performing an operation 2010-07-20 22:34:09 +03:00
kimchy 1ff8b43aa7 clean timeout listener from on going listeners once its removed 2010-07-20 21:11:56 +03:00
kimchy b3b5c9b0cc don't execute it on a thread pool 2010-07-20 18:34:01 +03:00
kimchy 553e8134ee create a send transaction exception on the calling thread and fill its stack, so we know where its coming from 2010-07-20 18:33:44 +03:00
kimchy 07480846c3 fix logging to actually pass the actual logging parameters 2010-07-20 18:33:12 +03:00
kimchy 6915b39dc3 clean timeout listener task when removing the listener 2010-07-20 18:32:49 +03:00
kimchy 45e54c1705 add retry if master is not connected till master is connected or a timeout has passed to master based operations (like cluster health) 2010-07-20 18:32:21 +03:00
kimchy 45c821316b no need for volatile on writer, its always accessed under a lock 2010-07-20 13:58:01 +03:00
kimchy 3c8cf68a17 improve timeout cluster service listener 2010-07-20 13:50:19 +03:00
kimchy 4ac1409afa don't list connect exceptions 2010-07-20 11:01:08 +03:00
kimchy 679f3758b7 ignore a more genric exception 2010-07-20 09:29:10 +03:00
kimchy 3867591ffa clean up the java api a bit 2010-07-20 09:18:16 +03:00
kimchy 2ee870e01e nicer logging, append space before index + shard id logging 2010-07-20 08:53:57 +03:00
kimchy 6f2c0f4e55 spelling 2010-07-19 17:51:42 +03:00
kimchy 5f0470d68b Support Cross-Origin resource in http/rest module, closes #218. 2010-07-19 15:06:13 +03:00
kimchy e48b1d98db more recovery under load tests, now with nodes shutdown, ignore node connected exception or retry when performing replicated operations 2010-07-19 11:33:21 +03:00
kimchy df602054fb fix relocation of primaries and perform the operation on the relocated primary replica as well 2010-07-19 09:54:52 +03:00
kimchy 0d20790ffe when shutting down a node, close shard (and indices) in parallel 2010-07-19 02:17:59 +03:00
kimchy d657d4447b more shard relocation tests, create a cached executor that can be used to allocate threads that are not affected by the actual thread pool impl (can always fork, no blocking) 2010-07-19 01:45:09 +03:00
kimchy fefcfb5b5c refactor recovery to be handled on the node level (and not per shard), with better retry mechanism when doing peer shard recovery 2010-07-18 22:54:21 +03:00
kimchy 8ac8dd818d now that we have a file based transaction log, increate the memory clean threshold 2010-07-18 21:24:52 +03:00
kimchy 49f8501a5b nicer serilization 2010-07-17 22:58:16 +03:00
kimchy 910ad2a408 fix optimization to not do sorting when searching with score "desc" sorting 2010-07-17 19:38:04 +03:00
kimchy 08ca383fd5 more work on cloud gateway 2010-07-17 15:20:13 +03:00
kimchy 6194b3ab0f add NaN handling for specific hit score as well 2010-07-17 13:52:31 +03:00
kimchy 1fccaf06e9 start work on reimplementing cloud gateway 2010-07-17 13:12:53 +03:00
kimchy e61dc78c21 max_score == NaN, closes #263. 2010-07-17 01:39:12 +03:00
kimchy 3006183191 fix a problem where max_score was NaN if there were no hits from a shard (though there are hits from other shards), it will still be NaN where there are no hits 2010-07-15 22:58:18 +03:00
kimchy cf998e2f89 protected when not of the same type mapper when merging 2010-07-15 22:35:18 +03:00
kimchy 9f6fde9adb better handling of failed merge because of different types 2010-07-15 20:40:40 +03:00
kimchy cfa56cb3d6 Put Mapping: When updating existing mappings, the request returns with acknowledged `false`, closes #262. 2010-07-15 20:03:44 +03:00
kimchy c987e8a534 Querying mapping on a non-master throws an error, closes #261. 2010-07-15 19:03:45 +03:00
kimchy ee60d7dedc better exception message when failing to parse a specific field 2010-07-15 18:46:12 +03:00
kimchy ca8ad83092 only count data_nodes with recover_after_nodes parameter 2010-07-15 01:12:00 +03:00
kimchy 61fadb4dc0 use safe method (to get proper exception) to get shard injector 2010-07-14 23:39:19 +03:00
kimchy f6fa056f1c fix REST api for cluster state and the ability to filter parts of the data returned 2010-07-14 16:05:47 +03:00
kimchy 95ba62f83d Translog: Implement a file system based translog and make it the default, closes #260. 2010-07-14 14:53:58 +03:00
kimchy e7d80b8244 remove max_shards_per_node setting (actually, you could not set it even...), will be revisted when proper SLA based shard allocation will be implemented 2010-07-14 14:29:56 +03:00
kimchy a6bd64f30d Facet results vary depending on size, closes #259. 2010-07-14 11:30:39 +03:00
kimchy 2704ab3d69 improve join process in cluster, fetch the cluster meta-data on join and handle new meta data 2010-07-14 10:34:22 +03:00
kimchy 141506afc5 improve join process in cluster, fetch the cluster meta-data on join and handle new meta data 2010-07-14 09:28:56 +03:00
kimchy 2e2f22fbd6 nullify the updater on stop 2010-07-14 08:55:57 +03:00
kimchy c4d6ccd07b nicer logging 2010-07-14 08:41:42 +03:00
kimchy 234455530a Make merging mappings smarter, closes #253. 2010-07-14 07:32:17 +03:00
kimchy 892dadca59 refactor meta data opertions into discrete services that do all operations within the cluster update process 2010-07-13 18:01:57 +03:00
kimchy ed04721f08 doc 2010-07-13 10:19:57 +03:00
kimchy f0cf552bc5 sorting breaks when sorting on a field that has no value in some documents 2010-07-13 09:38:55 +03:00
kimchy e6bd3f2693 Put Mapping: Fail when an analyzer is specified that was not configured, closes #252. 2010-07-12 23:18:32 +03:00
kimchy 65dffdf9cf throw a descriptive exception when trying to sort on multi token or multi values field per doc 2010-07-12 23:13:37 +03:00
kimchy ff17183798 add more sort tests 2010-07-12 22:25:49 +03:00
kimchy fe50a6f64e improve allocation of shards based on existing work location, wait with index creation on other nodes, so listing their respective existing location will work 2010-07-12 01:33:38 +03:00
kimchy 6722e05418 only list unallocated shards on data nodes 2010-07-11 22:11:06 +03:00
kimchy 52ac24fa23 more debug statr on recovery from gateway 2010-07-11 21:51:45 +03:00
kimchy e7a8da8236 do flush within the finalization of index shard recovery 2010-07-11 21:20:36 +03:00
kimchy 30b34b975c fix spelling 2010-07-11 20:53:03 +03:00
kimchy 294f09a1d7 Shutdown API: Improve behavior when shutting down the whole cluster, closes #250. 2010-07-11 20:41:58 +03:00
kimchy 09493691a7 nicer serializtion 2010-07-11 15:41:12 +03:00
kimchy 671a14ce87 freq is always positive int 2010-07-11 15:35:53 +03:00
kimchy a6ef805aa8 add doc 2010-07-11 00:40:29 +03:00
kimchy ef369fc40f javadoc 2010-07-10 23:21:34 +03:00
kimchy d0cdbeffba only send intiial state recived after sending join request, just when we actually get a new cluster state that includes us. Also, handle no nodes to send to with generic transport nodes action. 2010-07-10 15:43:49 +03:00
kimchy a0ead02299 Don't create / use the work directory if not needed (for example, on client / non data) nodes, closes #249. 2010-07-10 14:29:39 +03:00
kimchy 7ba3b3a9eb Zen Discovery: Control which nodes are allowed to become masters, closes #248. 2010-07-09 04:27:13 +03:00
kimchy b657ffc5e7 Zen Discovery: Control which nodes are allowed to become masters, closes #248. 2010-07-09 04:22:03 +03:00
kimchy d1f5577c40 add channel based read from fs snapshot, reusing the same file descriptor with direct position based reads 2010-07-09 03:51:36 +03:00
kimchy 6cdd6a015c reuse file buffer 2010-07-09 03:24:31 +03:00
kimchy d531d82cfb Zen Discovery: A node might get into an infinite state of trying to find a master (when client / non_master) nodes exists, closes #247. 2010-07-09 00:43:06 +03:00
kimchy 87eb6bed8f nicer logging 2010-07-08 22:53:41 +03:00
kimchy 720e6a6d5b nicer logging 2010-07-08 22:52:03 +03:00
kimchy be6aaa157f fixed a bug in fs translog where it wasn't seeking correctly, and not counting the number of operations 2010-07-08 22:18:26 +03:00
kimchy ef148077fc better logging 2010-07-08 21:37:10 +03:00
kimchy 84af9f5681 improve handling of translog snapshotting 2010-07-08 20:51:37 +03:00
kimchy 4429a61528 better cleanup on shard startup failure 2010-07-08 19:20:03 +03:00
kimchy b69fc265c5 index writer ram buffer size can only be 2G, make sure not to set an illegal value 2010-07-08 19:03:55 +03:00
kimchy 400ec7b8d1 index writer ram buffer size can only be 2G, make sure not to set an illegal value 2010-07-08 18:58:04 +03:00
kimchy 63deef42cf add a help appendable blob store based on immutable one 2010-07-08 18:53:40 +03:00
kimchy 8d669ff54e move sorting to be done using the new fielddata cache, and not using lucene built in FieldCache 2010-07-08 14:42:35 +03:00
kimchy 97b5b41522 enable eager cache eviction based on readers, requires a nasty hack to Lucene, basically overriding two classes to hook into when a reader actually gets closed (and only enabled when elasticsearch is before lucene in the classpath, which we take for in the startup scripts) 2010-07-07 17:15:26 +03:00
kimchy a36b06c682 ignore failed snapshot on a non started index 2010-07-07 16:59:33 +03:00
kimchy 5ff46911b6 ignore recovery when no node to recover from (it has disconnected, for example) 2010-07-07 16:30:28 +03:00
kimchy 3fb68c52da ignore snapshot on a shard thas has just been created 2010-07-07 16:30:17 +03:00
kimchy 11a8d637c0 ignore snapshot on a shard thas has just been created 2010-07-07 16:26:19 +03:00
kimchy e40aa540fe ignore recovery when no node to recover from (it has disconnected, for example) 2010-07-07 16:25:55 +03:00
kimchy 9578ad3ef1 Terms Facets: Allow to specify a set of terms to exclude in the request, closes #246. 2010-07-07 14:41:34 +03:00
kimchy 2b5458daf6 implement a simple file system based translog, not the defeault yet, should provide better memory management 2010-07-07 09:30:07 +03:00
kimchy b2bdb149f9 log shard close failure when closing an index 2010-07-06 22:50:40 +03:00
kimchy 6aa9be238c use the atomic integer counter to represent the size 2010-07-06 18:45:19 +03:00
kimchy 267016758d improve handling of memory caching with file system, only force compound file when really needed (when an extension that exists within the compound file is part of the memory cached extensions) 2010-07-06 18:21:36 +03:00
kimchy ac01bb86e8 clean lock only before opening an index writer 2010-07-06 18:13:05 +03:00
kimchy 2908418d37 checn the default byte buffer cache to 200mb, now that its global per node 2010-07-06 17:58:30 +03:00
kimchy 070cb5b295 allocate replica shards (backups) only after the primaries have been allocated, this results in better theoretical shard allocation, and the ability to reuse local storage index files (comared against the backup) 2010-07-06 17:58:05 +03:00
kimchy 1d39bb4d51 cache md5 computation for unallocated shard and reused it 2010-07-06 09:00:24 +03:00
kimchy d4f86899e3 more work on reuse work dir, improved transaction log to move to stream based 2010-07-06 00:12:40 +03:00
kimchy b078c9206a more work on reuse work dir, added md5 logic to identify what need to be recovered or not, note, requires reindexing 2010-07-05 14:13:55 +03:00
kimchy 79b69764e0 doc 2010-07-01 15:03:11 +03:00
kimchy e8b261c165 better shutdown handling 2010-07-01 10:28:57 +03:00
kimchy ad5525f575 fix unwrapping of exceptions 2010-07-01 10:27:53 +03:00
kimchy f6c9c49645 better logging when failing to read full content of a transport message 2010-07-01 03:00:57 +03:00
kimchy c75a56ca17 better shutdown handling 2010-07-01 02:41:00 +03:00
kimchy c2d02e4e3a wrong serialization of shardId 2010-07-01 02:18:35 +03:00
kimchy d848ab3df3 rename executors to esexecutors 2010-07-01 00:06:34 +03:00
kimchy b0d736b489 make the shard routing an actual component 2010-06-30 23:39:27 +03:00
kimchy 811559f393 add another test if the pool size is greater than the core pool size before a worker can exit 2010-06-30 23:08:05 +03:00
kimchy 6709518c4e add the ability to list shard index blobs without actually having one created 2010-06-30 22:54:59 +03:00
kimchy 66096e97f2 Remove the terms API (replaced by facets in search), closes #242. 2010-06-30 19:49:41 +03:00
kimchy df7e3f55a4 improve cluster blocks data structures for simpler usage 2010-06-30 16:06:31 +03:00
kimchy 4049da97fb time value to support parsing days as well 2010-06-30 13:00:25 +03:00
kimchy 088e0b5a64 Automatic management of indexing buffer size, closes #241. 2010-06-30 01:46:57 +03:00
kimchy fdb2eff998 Automatic management of indexing buffer size, closes #241. 2010-06-29 22:54:54 +03:00
kimchy 3f6cd46736 allow to configure tcp network modules (transport, memcached and http) to use blocking io (just enabling it to do some performance tests) 2010-06-29 21:07:38 +03:00
kimchy 01ff81fa89 fix log message 2010-06-29 19:49:28 +03:00
kimchy bc803b4fcf close the thread pool on test 2010-06-29 18:35:03 +03:00
kimchy 703657699a refactor new thread pools in, remove dynamic ones 2010-06-29 18:20:31 +03:00
kimchy de4b21e986 backups don't become primaries on node failures, primaries are first allocated and then start initializing from gateway 2010-06-29 18:20:13 +03:00
kimchy 514df4ee3f refactor the shard routing abstraction, no need for it 2010-06-29 18:02:10 +03:00
kimchy fe5e5a073b work on scaling and blocking thread pool 2010-06-29 18:01:54 +03:00
kimchy 9051ae3a65 work on scaling and blocking thread pool 2010-06-29 17:50:05 +03:00
kimchy 297001de53 update benchmark code 2010-06-29 08:59:28 +03:00
kimchy aa5791a14f move block metadata checks to the transport level 2010-06-29 08:51:36 +03:00
kimchy a82bf74c1c expose blocks in the cluster state rest api 2010-06-29 01:00:07 +03:00
kimchy 5f538b1ba3 Block operation performed on cluster until it recovered from the gateway, closes #239. 2010-06-28 22:53:13 +03:00
kimchy b970f0dcc2 fix toString of bytes value 2010-06-28 17:53:07 +03:00
kimchy 63523f4525 Query DSL: field query does not take into account `allow_leading_wildcards`, closes #236. 2010-06-28 12:33:36 +03:00
kimchy cb9be9973b add the ability to define meta _attributes for mapping, basically a place to store additional unstructured data on the mapping 2010-06-28 09:10:08 +03:00
kimchy 3770924300 allow to control the buffer size of the gateway blob container 2010-06-27 23:26:32 +03:00
kimchy a7c13826da more work on reusing work directory, clean unallocated shards when they are not needed 2010-06-27 18:19:12 +03:00
kimchy b43f0f5965 add streamable to stores files metadata 2010-06-27 14:18:14 +03:00
kimchy 4ae45946a2 nicer logging 2010-06-27 13:50:27 +03:00
kimchy fc7392a98b fix logging 2010-06-27 13:07:21 +03:00
kimchy 90c995dda5 add todo 2010-06-27 12:28:30 +03:00
kimchy 1e7481d5ab fix logging 2010-06-27 12:15:58 +03:00
kimchy c56609a0ad fix default merge size to not truncate on long 2010-06-27 12:14:38 +03:00
kimchy bbb9c3883f fix jdoc 2010-06-27 12:05:30 +03:00
kimchy f1f695b4d9 fix logging 2010-06-27 09:39:07 +03:00
kimchy b9590b5ed6 fix logging 2010-06-27 09:37:44 +03:00
kimchy edd8786a81 close open index input on failure 2010-06-27 02:13:59 +03:00
kimchy 33d357dbb4 Memory Store / FS Memory: Create a node level memory store cache and allocator, closes #235. 2010-06-26 22:34:30 +03:00
kimchy 00d2abef40 refactor facets builder API to make it more manageable 2010-06-26 01:36:42 +03:00
kimchy 5df2257319 add test for facets with search size 0 2010-06-25 22:46:07 +03:00
kimchy 83010f7ee1 more apis on the index level store to help clean unused work shards when needed 2010-06-25 03:39:42 +03:00
kimchy 811856e1f8 better method name 2010-06-25 01:41:06 +03:00
kimchy 98e93322ae use java user.dir as the default "path.home" location 2010-06-24 22:11:13 +03:00
kimchy 4709f00790 add to index store the ability to get the backing store free / total space 2010-06-24 18:05:21 +03:00
kimchy edf0075025 more work on reusing work node level data, now when recovering from peers 2010-06-24 15:34:02 +03:00
kimchy 57169d4233 initial work on reusing work node level data when recovering from gateway 2010-06-24 00:32:05 +03:00
kimchy 296b6b2f57 use custom similarity in search (if there is one) 2010-06-23 14:30:14 +03:00
kimchy a7fdd36f6e clean files that are in the store but not in the gateway 2010-06-23 12:21:04 +03:00
kimchy 2121aaad3b part of the index store enhacement 2010-06-23 10:49:06 +03:00
kimchy c4aff14e15 remove field options from field cache and freqs, simpler and they are not used in any case 2010-06-23 09:48:46 +03:00
kimchy cccce2b114 fix work with cluster setting 2010-06-23 08:00:29 +03:00
kimchy 3b4584040b add index level store constuct (on top of the shard level store) 2010-06-23 07:47:18 +03:00
kimchy 8e62bb6934 Cluster State API: Allow to filter the state on nodes, routing_table, metadata, and indices, closes #234. 2010-06-22 16:19:10 +03:00
kimchy dcb0d18519 javadoc 2010-06-22 16:09:49 +03:00
kimchy 622d4041b8 handle cases where the node does not exists in the disco nodes 2010-06-22 10:59:24 +03:00
kimchy 7c931f34fa ignore state exceptions when trying to flush 2010-06-22 10:26:44 +03:00
kimchy 0c6b82aeca fix javadoc 2010-06-22 09:26:37 +03:00
kimchy 3d84af2a40 refactor hdfs gateway to use the new common blobstore 2010-06-22 09:10:00 +03:00
kimchy 21627bca10 return after notifiying listener on failure 2010-06-21 23:29:13 +03:00
kimchy 893de86e9e fix missing break statemetns (via Lukas) 2010-06-21 22:45:02 +03:00
kimchy 7ed7c6db4e Gateway: Internal refactoring, requires manual upgrade when using fs gateway, closes #232. 2010-06-21 22:40:12 +03:00
kimchy 733abdcac3 more unboxed math 2010-06-20 14:42:11 +03:00
kimchy 701920c6c1 remove unused API in field mapper (that also cause compilation to break on eclipse) 2010-06-20 12:12:01 +03:00
kimchy e846ed6465 Lucene: Upgrade to latest 3.0.2 version, closes #231. 2010-06-20 01:04:39 +03:00
kimchy 384f8a4f42 Return the maxScore per search and score for each search hit, closes #205. 2010-06-20 00:23:27 +03:00
kimchy 0f2147aeec add some prepare APIs to the groovy APIs 2010-06-19 21:09:03 +03:00
kimchy 891c69f8de refactor builder requets into common base class 2010-06-18 14:26:33 +03:00
kimchy adef9d69ff Gateway: Allow to configure a `recovery_after_time` and `recover_after_nodes`, closes #223. 2010-06-18 05:32:51 +03:00
kimchy 22926f0026 Analysis: When specifying empty array for stopwords, use an empty list for stopwords, closes #230. 2010-06-18 00:34:18 +03:00
kimchy 8e4c139c8b Analysis: fields that are `not_analyzed` should automatically default to keyword analyzer, closes #229. 2010-06-17 23:23:53 +03:00
kimchy 60bd8f34d2 add debug logging of the ping responses received 2010-06-17 22:37:47 +03:00
kimchy f64b96e3c8 add an exception if there is no master node set 2010-06-17 22:19:43 +03:00
kimchy 7b145fd4e1 Expose the mappings of a given index or index/type, closes #222. Added the ability to filter returned data from the cluster state API, and use it to expose REST GET endpoint for "/_mapping", "/{index}/_mapping", "/{index}/{type}/_mapping" ({index} can also be "_all"). 2010-06-17 17:56:07 +03:00
kimchy 3bc8c307f1 simplify netty transport to use single channel 2010-06-17 12:50:01 +03:00
kimchy 597f020156 fork node connected and disconnected on a different thread pool 2010-06-17 12:47:42 +03:00
kimchy 662ded3801 connect to a node when it joins the cluster on the disco level, so if it fails, it will be propagated back and the node will not be added to the cluster state 2010-06-17 12:46:53 +03:00
kimchy 86134d32a8 nicer logging 2010-06-17 12:46:05 +03:00
kimchy d1c7bde07e make the rest cluster state return mappings as object keyed type 2010-06-17 02:45:17 +03:00
kimchy 38005826e7 detect write errors early and notify transport handler 2010-06-17 02:17:54 +03:00
kimchy 8af925a0b1 add internal flags to help simulate forcefull disconnections 2010-06-17 01:05:23 +03:00
kimchy a60e542718 cancel the timeout if we fail to send the request 2010-06-16 23:07:41 +03:00
kimchy b0c8255cd9 clean the cluster state when starting (handles restarts better) 2010-06-16 21:59:26 +03:00
kimchy a59679b0c6 better failure messages 2010-06-16 21:47:58 +03:00
kimchy 4d26acd6da add aliases to cluster state rest response 2010-06-16 06:08:36 +03:00
kimchy a547c1ccc0 cluster state to return native json (if possible) as part of the mappings 2010-06-16 05:48:30 +03:00
kimchy 626989c55e don't return the source when asking for scripted fields 2010-06-16 04:48:06 +03:00
kimchy 2d6f61b3c1 Facets: Script Histogram facet, closes #228. 2010-06-16 04:22:00 +03:00
kimchy bb24b56c66 Facets: Script statistical facets, closes #227. 2010-06-16 02:18:24 +03:00
kimchy c2786038e2 Script Filter: Support providing a custom script as a filter, closes #226. 2010-06-16 01:32:04 +03:00
kimchy be3b779caa Tcp Transport: Reduce `transport.tcp.connection_per_node` from 5 to 1, closes #225. 2010-06-15 23:36:31 +03:00
kimchy 8e589b5a04 nicer code 2010-06-15 23:14:04 +03:00
kimchy eaf124123d reduce long name 2010-06-15 23:09:09 +03:00
kimchy 65039ff21d Zen Discovery: When a master node is forcefully killed, other nodes might not monitor the other elected master, closes #224 2010-06-15 23:08:52 +03:00
kimchy eb68891ae5 move size value/unit to bytes size value/unit, and create a ten based size value/unit 2010-06-15 18:26:44 +03:00
kimchy a18e58b394 move size value/unit to bytes size value/unit, and create a ten based size value/unit 2010-06-15 18:16:10 +03:00
kimchy f75abbaa00 big refactoring thanks to proper jarjar built from source with asm 3.3 allowing to jarjar guice and others, includes package relocations 2010-06-15 17:59:04 +03:00
kimchy d5bd77cc38 big refactoring thanks to proper jarjar built from source with asm 3.3 allowing to jarjar guice and others, includes package relocations 2010-06-15 17:28:05 +03:00
kimchy 636bfe8466 big refactoring thanks to proper jarjar built from source with asm 3.3 allowing to jarjar guice and others, includes package relocations 2010-06-15 16:51:38 +03:00
kimchy 48979ab6c8 improve gateway recovery when using delay index creation, close the loophole when the master was shut down before the delay expired and now other node becoming master will do the recovery 2010-06-15 10:27:26 +03:00
kimchy 98df1b3433 Gateway: Allow to configure a delay till index creation from gateway will occur, closes #223. 2010-06-15 01:25:55 +03:00
kimchy e0d20af743 ignore when flush can't be done 2010-06-14 13:25:18 +03:00
kimchy 0a1bc874c3 Search: Allow to specify script fields to be loaded, closes #221. 2010-06-14 12:50:33 +03:00
kimchy c087bbe804 use List for fieldNames to reduce extra array construction 2010-06-14 04:36:03 +03:00
kimchy 953779ccea allow to pass parameters to custom score script 2010-06-14 04:12:57 +03:00
kimchy 1b32c1ccf4 add cleanable thread local to single values 2010-06-14 03:42:43 +03:00
kimchy 751c56f31e Query DSL: custom score (scripted) query, closes #220. 2010-06-14 03:15:23 +03:00
kimchy 2d0785078c move to field.data package 2010-06-13 18:05:59 +03:00
kimchy a9fc276a3e refactor the custom boost factor query into a more general function boost query 2010-06-13 16:51:19 +03:00
kimchy ec481159d6 pass score docs in order in custom boost factor query 2010-06-13 12:41:16 +03:00
kimchy 2daa54a2bb don't clean thread locals explicitly, let GC clean them 2010-06-12 12:45:38 +03:00
kimchy d515fe8644 add internal support for Rest/HTTP OPTIONS and HEAD 2010-06-11 00:13:08 +03:00
kimchy 5205a183e8 use deletion aware constant score query 2010-06-10 15:04:36 +03:00
kimchy 668e84a99e add specific doc object, DocFieldData and specific types extension to allow in the future for simpler doc level operations 2010-06-09 23:35:31 +03:00
kimchy 497b4a4f10 filtres should implement equals and hashcode 2010-06-09 21:23:19 +03:00
kimchy 64479a11c3 Search Facets: Histogram facets to allow to specify different key and value fields 2010-06-09 00:50:07 +03:00
kimchy 5ca050ffcb Search Facets: Facets to allow to define filters on them, closes #217. 2010-06-08 22:30:21 +03:00
kimchy 6257b78243 Query DSL: Add `and`, `or`, and `not` filters, closes #216 2010-06-08 18:04:34 +03:00
kimchy 47b3a81bec Search Facets: Histogram Facets, closes #215 2010-06-08 15:45:58 +03:00
kimchy d36cea07f3 refactor some facet support 2010-06-08 11:46:30 +03:00
kimchy 874993557c add sum of squares, variance, and std deviation for statistical facet 2010-06-08 10:39:45 +03:00
kimchy 26ed029d8a Query DSL: Support fuzzy query, closes #213. 2010-06-07 18:13:57 +03:00
kimchy c772e7c337 better comment 2010-06-06 14:18:58 +03:00
kimchy f0c56ce18e nicer code, internalize Facets and create a clean external interface 2010-06-06 03:14:34 +03:00
kimchy 81deb833de numeric statistical facets support 2010-06-05 23:58:05 +03:00
kimchy 8aeb589a42 numeric statistical facets support 2010-06-05 23:44:34 +03:00
kimchy 57a21bedd2 update doc 2010-06-05 21:59:42 +03:00
kimchy 88ba980f79 refactoring of facets to be more open for different facet types 2010-06-05 21:49:10 +03:00
kimchy d9d9304f47 add TODO 2010-06-05 15:59:06 +03:00
kimchy 83ad15b9f2 improve cleaning algo, force clean only ones that have not been cleaned, and do a force clean only when we start reaching the full clean 2010-06-05 14:35:12 +03:00
kimchy 5cdba0383b Flush API: Allow to provide `full` parameter for a complete clean, closes #210. 2010-06-05 06:10:02 +03:00
kimchy ef1866aed4 add flush logging 2010-06-05 03:10:55 +03:00
kimchy bcbc0dd741 Search: Search requests hangs when no indices exists, closes #209. 2010-06-05 01:36:24 +03:00
kimchy d0eb836c4a Plugins: Allow to easily plug a custom DSL query/filter parsers, closes #208. 2010-06-05 01:07:12 +03:00
kimchy 9f8644472e more tuning of field data capabilities, make term facet collector handle any type (though it really make sense mainly for string types) 2010-06-04 14:03:43 +03:00
kimchy 0901322337 rename field facet to term facet 2010-06-04 07:03:30 +03:00
kimchy 6e15e3dd24 add helper methods to field data 2010-06-04 06:15:35 +03:00
kimchy 13d7f10b7b add helper methods to field data 2010-06-04 04:42:44 +03:00
kimchy 66c9f2f834 Search Facets: Field Facets, closes #207 2010-06-04 02:31:49 +03:00
kimchy dbb7df742e protected if the cached stream got evicted on us 2010-06-03 20:48:57 +03:00
kimchy c68787aa4c most master node outside of the nodes section, so all elements in nodes will have the same strucutre 2010-06-03 14:30:54 +03:00
kimchy d9ec629473 clean log 2010-06-03 11:42:20 +03:00
kimchy b2f90a2133 REST API does not expose node-master status, closes #203. 2010-06-02 17:04:43 +03:00
kimchy 6c8f49c37d REST API does not expose node-master status, closes #203. 2010-06-02 16:59:55 +03:00
kimchy c2d583064e FS Gateway: Allow to configure is native file copying will be used, closes #202. 2010-06-02 12:06:41 +03:00
kimchy 5ef421e779 Nodes Info API: Failed to generate REST response when node attributes are set, closes #201. 2010-06-02 01:10:43 +03:00
kimchy a7ad295f63 Zen Discovery: ungraceful shutdown of the master and start of replacement node might cause the cluster not to elect a new master, closes #200. 2010-06-01 22:29:38 +03:00
kimchy ec662cc019 always start the unicast ping discovery, so unicast discovery will work even when using multicast 2010-05-31 22:24:51 +03:00
kimchy 4b85407f5e work on a nicer java api 2010-05-29 22:23:39 +03:00
kimchy e59b413980 work on a nicer java api 2010-05-29 05:50:17 +03:00
kimchy 3406e77cec Streamline single value with boost queries: Allow for both value and query name to specify the query value, closes #192. 2010-05-28 12:23:51 +03:00
kimchy 493464ba02 improve logging 2010-05-28 05:26:51 +03:00
kimchy a7cce9cd38 Replication Actions: Allow to control replication type - `async` or `sync`, closes #196. 2010-05-28 03:56:04 +03:00
kimchy 84a5c1eac8 Replication Actions: Allow to control replication type - `async` or `sync`, closes #196. 2010-05-28 03:47:35 +03:00
kimchy 477024e3be cached termsSet used in dfs phase 2010-05-27 20:31:48 +03:00
kimchy 7e5bfb5d3b share the open channels handler 2010-05-27 18:25:59 +03:00
kimchy ded270fa03 add search id when failing to execute 2010-05-27 14:29:04 +03:00
kimchy 71ece4e31c rever back all interfaces binding in multicast 2010-05-27 13:12:33 +03:00
kimchy d7984e9962 rever back all interfaces binding in multicast 2010-05-27 12:43:05 +03:00
kimchy d14a8afaf0 add another trace logging 2010-05-27 12:12:18 +03:00
kimchy 101652d720 add another trace logging 2010-05-27 12:10:55 +03:00
kimchy 74fa3ed32f Zen Discovery: Improve Multicast Binding and Sending, closes #195. 2010-05-27 02:11:06 +03:00
kimchy 9433f4d651 Network Settings: Allow to use '_' as well as '#' for special host names, closes #194. 2010-05-26 16:33:35 +03:00
kimchy 40d4a350fd nicer code 2010-05-26 13:01:16 +03:00
kimchy 26364afd7e Failed to acquire lock, closes #193. 2010-05-26 12:29:13 +03:00
kimchy 463910da45 handle null doc id set 2010-05-26 11:20:15 +03:00
kimchy 9a49629d17 throw an exception when an action is not found 2010-05-26 09:16:10 +03:00
kimchy 32e4c405de add async http client 2010-05-24 23:27:07 +03:00
kimchy b7d11f1303 Gateway Storage: Improve it to support non breaking changes in the future, closes #191. 2010-05-24 18:27:48 +03:00
kimchy 116cfce6f2 FS Gateway: Change indices location (Requires manual upgrade), closes #188. 2010-05-23 08:37:41 +03:00
kimchy 5c6864e8b3 inital work on hadoop plugin - gateway hdfs support 2010-05-23 08:17:52 +03:00
kimchy 596f9db8d8 fix to build the facets map on getFacets 2010-05-22 21:12:07 +03:00
kimchy a372643ca9 remove using Lucene StringHelper in the code 2010-05-22 16:52:22 +03:00
kimchy 4c51ddd2ad update comment for next lucene version 2010-05-22 13:22:27 +03:00
kimchy 66158ffc19 fix logging 2010-05-22 02:50:35 +03:00
kimchy 6191bf9cdc Optimize API: Change flush and refresh to default to true and not false, closes #187. 2010-05-22 02:38:38 +03:00
kimchy 10a245e76d fix problem with transport client starting 2010-05-22 02:13:39 +03:00
kimchy bcc59ac4cc comment for next lucene version 2010-05-22 02:00:26 +03:00
kimchy ecc74f225e Cloud Plugin: Cloud gateway default chunk size change to 1g, closes #186. 2010-05-22 01:08:43 +03:00
kimchy fb03652ecd another simple test 2010-05-21 20:40:10 +03:00
kimchy dc4acde3db Network Settings: Default publish host to first non loopback (first ipv4 then ipv6), closes 185. 2010-05-20 22:31:45 +03:00
kimchy b4940d258c fix FD with zen discovery to actually retry and not rely on closed socket 2010-05-20 22:05:47 +03:00
kimchy 3e0f854c88 add network info printing 2010-05-20 21:10:53 +03:00
kimchy a8a4bbc30e internal optimization - share built in analyzers on the node level and not on index level 2010-05-20 00:55:31 +03:00
kimchy 5718e27fcf add clean index to indices 2010-05-19 22:00:27 +03:00
kimchy d1acef1e09 Filter Cache: Improved Caching, closes #182. 2010-05-19 21:05:43 +03:00
kimchy bd6b89f7ca Query DSL: Allow to control if filters should be cached, closes #181. 2010-05-19 20:41:29 +03:00
kimchy 1882460645 add an index level engine instance, allowing to provide and share data across different shard level engines 2010-05-19 04:24:39 +03:00
kimchy 40fdcc43b7 update all to use LUCENE MONITOR 2010-05-19 03:55:07 +03:00
kimchy 7639f9b784 change the addition of analysis processors to override the default ones 2010-05-19 01:40:34 +03:00
kimchy b5140085a5 better logging 2010-05-19 00:45:13 +03:00
kimchy 10bc248000 add anotehr test 2010-05-18 18:23:01 +03:00
kimchy cb5500919a add filter cache tests 2010-05-18 17:47:19 +03:00
kimchy 03d7cf3171 remove dead code 2010-05-18 14:55:00 +03:00
kimchy b7e540825f add seek back on mem directory 2010-05-18 04:49:39 +03:00
kimchy eaa75a70c7 add index writer tests 2010-05-18 04:43:46 +03:00
kimchy dc5bb20331 no need to cache the analyzer providers 2010-05-18 02:49:17 +03:00
kimchy 4ba2453369 create a single place where Lucene version is defined 2010-05-18 02:44:50 +03:00
kimchy 9ebb95c6c0 add common http settings 2010-05-18 01:35:35 +03:00
kimchy e696b4ca00 add common tcp settings 2010-05-18 01:22:59 +03:00
kimchy 34103c6b7f test and start of node stress tests 2010-05-17 23:56:42 +03:00
kimchy eef6c02bf7 test and start of node stress tests 2010-05-17 23:27:28 +03:00
kimchy 8d3347bb5c Memory Monitor: Remove explicit GC call, clear cached instead, closes #179. 2010-05-17 19:02:35 +03:00
kimchy 45ae8f6a95 lighter search holder 2010-05-17 17:50:32 +03:00
kimchy 23d2799d71 uppercase first letter in property name breaks "_all" queries, closes #178. 2010-05-17 17:23:24 +03:00
kimchy dbbb341136 better voaltile management 2010-05-17 01:27:41 +03:00
kimchy a8b9b01f3d correct wordings 2010-05-17 00:08:11 +03:00
kimchy 918e2a6882 don't increment the counter twice 2010-05-16 23:50:39 +03:00
kimchy ab57fa86af Zen Discovery: Increase Ping timeouts, closes #177. 2010-05-16 23:40:55 +03:00
kimchy 216dda3f9c Throttling of recovery (both gateway recovery and peer node recovery), closes #176. 2010-05-16 23:37:56 +03:00
kimchy 2301b0b13d chagne the defualt reader cleaner scheduler to 10 seconds 2010-05-16 20:06:59 +03:00
kimchy 013e7699c3 Index Request wrongly stores extra bytes as part of the document (trailing null bytes), closes #175. 2010-05-16 17:56:52 +03:00
kimchy 7b1aac5bb4 eagerly clean cache when shard gets closed 2010-05-16 03:57:26 +03:00
kimchy 5e4343b48b change how filter cache works, do weak/soft on IndexReader key, so we release the IndexReader on memory constraints 2010-05-16 03:29:56 +03:00
kimchy 5fa3e40a4e move to use simple CHM instead of the non blocking one 2010-05-16 03:29:05 +03:00
kimchy eef24a7b1f nicer log output 2010-05-16 01:31:23 +03:00
kimchy 2a79ffdc00 no need for explicit array creation 2010-05-16 01:23:50 +03:00
kimchy dfab82272a static classes 2010-05-16 01:15:10 +03:00
kimchy e5737e3058 bette logging 2010-05-16 01:09:50 +03:00
kimchy a005dc2c1f add transport info and stats to node info / stats api 2010-05-15 19:49:27 +03:00
kimchy b816ed5cf2 add thread pool stats and info to nodes stats / info api 2010-05-15 18:10:24 +03:00
kimchy 723e47a54b Consolidate network settings, common tcp settings, and generalized port settings, closes #174. 2010-05-15 16:30:02 +03:00
kimchy c9be7bde52 Network Settings: Add `host` setting that automatically set both `bind_host` and `publish_host`, closes #173. 2010-05-14 19:22:56 +03:00
kimchy a7f9db64fc add '-' option before install and remove 2010-05-14 13:43:21 +03:00
kimchy 04cfacec41 add tests for path mappings 2010-05-14 12:26:41 +03:00
kimchy 911d800ff8 add tests for path mappings 2010-05-14 12:17:17 +03:00
kimchy 10754b0873 better safeguard for concurrent restart requests 2010-05-14 10:28:18 +03:00
kimchy b49864299d fix wrong optimization on boolean query 2010-05-14 08:16:10 +03:00
kimchy f6d4654daa use ip address if possible when serializing inet socket transport information 2010-05-14 00:23:18 +03:00
kimchy d964f10bbe use ip address if possible when serializing inet socket transport information 2010-05-13 23:14:22 +03:00
kimchy 3d310406b7 fix header doc 2010-05-13 21:07:30 +03:00
kimchy 6cd3fc92ed reduce bytes allocation when doing http 2010-05-13 20:42:59 +03:00
kimchy fbde2c84ac guard for several restart requests for the same node at the same time 2010-05-13 20:14:01 +03:00
kimchy 8573c26602 improve mlt serialization and deserialization 2010-05-13 18:50:34 +03:00
kimchy 836461e6de improve search serialization and deserialization 2010-05-13 18:39:24 +03:00
kimchy bbdbfbeb59 improve count serialization and deserialization 2010-05-13 18:12:30 +03:00
kimchy a9cac052ec remove using deprecated method 2010-05-13 17:00:16 +03:00
kimchy 6a20ca562c more hardening of cleaning out buffers in case of failure 2010-05-13 16:52:59 +03:00
kimchy d2bc6ace83 when restarting, the latest cluster state should be cleared 2010-05-13 15:39:55 +03:00
kimchy e87a4117b6 expose settings builder in the node builder API 2010-05-13 15:17:49 +03:00
kimchy f6777f1e03 jarjar netty as well, so it won't be exposed to users using elasticsearch embedded 2010-05-13 14:41:50 +03:00
kimchy 7ba1ad2d7b create shard count request with the actual count request 2010-05-13 09:13:21 +03:00
kimchy b0dd31e600 better delete by query request serialization handling 2010-05-13 09:02:59 +03:00
kimchy 0e163e3649 better delete by query request serialization handling 2010-05-13 09:00:25 +03:00
kimchy e35d67afd7 better index request serialization handling 2010-05-13 08:43:09 +03:00
kimchy 4d6f2d56f0 Transport: A failure to handle a response might cause the transport to stop working, closes #170. 2010-05-12 21:59:58 +03:00
kimchy a11c7beef8 add debug logging 2010-05-12 17:27:50 +03:00
kimchy 5b8dac3dd7 fix typos 2010-05-12 16:57:31 +03:00
kimchy 88badbd52f fix javadoc 2010-05-12 16:50:01 +03:00
kimchy 7edb3f8d54 nicer code 2010-05-12 16:48:22 +03:00
kimchy 02c8706c33 better log message 2010-05-12 15:52:01 +03:00
kimchy eaa32f28b3 reset the master ping counter when we get a proper response 2010-05-12 15:50:15 +03:00
kimchy a0b25ec4c3 abstract non blocking hashset 2010-05-12 04:06:44 +03:00
kimchy f6509930c7 upgrade to netty 3.2.0 CR1 2010-05-12 04:03:20 +03:00
kimchy fdf6eb9cb2 Mapper: Date format - allow for multiple formats using '||' separator, closes #169. 2010-05-12 03:32:21 +03:00
kimchy 8df704e605 allow for json and xson to be parsed in content type 2010-05-11 22:40:02 +03:00
kimchy 7c53ef1084 use offset when detecting xcontent type 2010-05-11 22:38:00 +03:00
kimchy 040030dac8 reafctor inject to util from util.guice 2010-05-10 01:34:22 +03:00
kimchy 7721b86995 don't let guice by mistake inject two instances 2010-05-10 01:15:08 +03:00
kimchy 35c8774304 start work on a more scalable thread pool, expose its status over jmx 2010-05-10 01:13:16 +03:00
kimchy 7fe5243517 share cached bytes 2010-05-09 20:11:54 +03:00
kimchy 3e405c3ec7 Nodes Stats API, closes #165, 2010-05-09 09:59:43 +03:00
kimchy 39cb08fc1c abstract using non blocking long map into an interface, with an implementation of juc CHM as well 2010-05-09 07:14:02 +03:00
kimchy 462d426c53 expose more information using the node info api 2010-05-08 23:01:38 +03:00
kimchy 6424d02be8 more work on plugin script 2010-05-08 21:18:00 +03:00
kimchy 9872b4977d cache data stored in http 2010-05-08 03:34:31 +03:00
kimchy 31d226188f REST API: Allow to provide `case` parameter, with `camelCase` to return results in CamelCasing, closes #162. 2010-05-08 00:01:36 +03:00
kimchy e222d38b84 more work on monitoring support 2010-05-07 22:49:36 +03:00
kimchy 4ed93d20f6 more work on monitoring support 2010-05-07 19:16:08 +03:00
kimchy c05e2ad1f1 Query DSL: Custom Boost Factory Query, closes #161. 2010-05-07 01:45:30 +03:00
kimchy b54f6b77de add os level stats and info (still some more info left...) 2010-05-06 23:46:53 +03:00
kimchy 371bfebd16 initial support for process monitoring using sigar 2010-05-06 21:07:24 +03:00
kimchy 5603c4d09c add the ability to control the content type the map will serialize under 2010-05-06 15:50:35 +03:00
kimchy 2605bf7e89 add unicast ping test 2010-05-06 09:21:56 +03:00
kimchy 66f011ae76 allow to filter out settings from REST serivces 2010-05-06 09:06:32 +03:00
kimchy db4afc8750 Mapper: Object mapper default 'date_formats' to also support 'yyyy/MM/dd HH:mm:ss' and 'yyyy/MM/dd', closes #159. 2010-05-05 19:45:54 +03:00
kimchy 71d4f1ce7a seems like jackson does not support this feature yet... 2010-05-05 17:54:03 +03:00
kimchy 5b5a95ea77 better implemenation of InputStream by different StreamInput 2010-05-05 10:24:36 +03:00
kimchy 5b231c68f8 move to use LinkedBlockingQueue instead of LinkedTransferQueue so we eagerly clear memory when a new translog is created 2010-05-05 01:17:46 +03:00
kimchy 4ead4f5855 fix failed compilation 2010-05-05 00:34:59 +03:00
kimchy 8053237643 move from gcommon package 2010-05-04 16:02:26 +03:00
kimchy 24ccd73c5d add '/' to auto detection of dates 2010-05-04 15:26:59 +03:00