b27cc1318f
streams to handle BytesReference as a generic type as well
2012-07-08 01:59:24 +02:00
12beff9176
use netty frame decoder again
...
with the new bytes reference abstraction, and the new composite buffer handling in frame decoder, it makes more sense to move back and use netty frame decoder again, since we can easily slice refernces acorss composite buffers, and reduce copies when using compression
2012-07-08 00:56:56 +02:00
f7b538e17f
move rest request to provide content as BytesReference
...
this allows for better buffer usage, specifically when forwarding requests to other nodes
2012-07-07 22:49:31 +02:00
5f5458fd56
add byte buffer backed bytes reference
2012-07-07 21:33:01 +02:00
35233564fd
buffer management refactoring
...
First phase at improving buffer management and reducing even further buffer copies. Introduce a BytesReference abstraction, allowing to more easily slice and "read/write references" from streams. This is the foundation for later using it to create smarter buffers on top of composite netty channels for example (which http now produces) as well as reducing buffer copies when sending transport/rest responses.
2012-07-07 01:26:41 +02:00
5d93b2bfe9
Thread Pool: Allow to configure the query_type for fixed thread pool (linked/array), defaults to array, closes #2086 .
2012-07-06 11:48:45 +02:00
8d1e04a973
have the quick rolling restart stress test also wait for 0 relocating shards
2012-07-06 01:01:18 +02:00
dcf9194820
fix naming, also make explainable search script extend search script
2012-07-06 01:00:44 +02:00
a77c824476
Add an interface to be implemented by native scripts to be able to explain the score
2012-07-06 00:55:12 +02:00
29826863dd
Local Gateway: old global state files are not properly deleted, closes #2085 .
2012-07-06 00:19:34 +02:00
d844093953
only set the decoder type if set in the settings
2012-07-05 23:47:03 +02:00
2328776ec5
allow to configure the decoder type for LZF, between optimal and safe
2012-07-05 23:46:02 +02:00
0a615ca35a
when needing to copy over the bytes from the netty content channel, zero it out so ti can be GC'ed
2012-07-05 23:36:27 +02:00
5f1b1c6f69
Upgrade to Netty 3.5.2, closes #2084 .
2012-07-05 23:09:26 +02:00
57023c8ba9
Compression: Support snappy as a compression option, closes #2081 .
2012-07-04 17:14:12 +02:00
2436f645aa
add 0.19.9
2012-07-02 22:10:28 +02:00
1c3c737524
don't double compress already compressed files when doing recovery
2012-07-02 21:41:40 +02:00
1668533556
improve dangling index support to not detect explicit deleted index as dangling, harden when we delete the _state of an index
2012-07-02 17:17:16 +02:00
b2c4876626
(full) recovery memory based indices with local gateway, don't fail them, closes #2077 .
2012-07-02 00:31:10 +02:00
bdf146ded6
Exists API can send response body, closes #2072 .
2012-07-02 00:04:49 +02:00
3ab85bacdd
don't delete shards state in the state handle, but let the indices store eventually delete it based on its logic of when to delete a shard
2012-07-01 21:51:19 +02:00
6cd9b92f95
ElasticSearchIllegalStateException: stream marked as compressed, but no compressor found on transport layer, closes #2076 .
2012-07-01 20:26:33 +02:00
743d3f7e4b
better failure message when failing to detect compressor on compressed transport stream
2012-07-01 18:34:58 +02:00
e5c89def42
Support wildcard and +/- notation for multi index APIs, closes #2074 .
2012-07-01 18:16:04 +02:00
22bdd161ad
return proper content type for mp3
2012-06-30 18:26:21 +02:00
565db26e13
Store Compression: integer overflow causes failed reads (index is safe), closes #2071 .
2012-06-30 01:37:46 +02:00
8bab859822
simplify tests, doc file length
2012-06-29 16:01:17 +02:00
f2e39e4ee2
Auto import dangling indices, closes #2067 .
2012-06-29 01:01:26 +02:00
07454243e3
getSourceAsString() for doc inserted as SMILE fails, auto convert to JSON, closes #2064 .
2012-06-28 14:17:58 +02:00
a872c88f03
dangling index handling might still remove the state files for the dangling index, closes #2065 .
2012-06-28 13:32:44 +02:00
7454c7c192
freeing the search context should execute on the same thread, its cheap
2012-06-28 09:56:49 +02:00
016e2e7288
Have rest request also allow to get the content as a streamable
...
(note, this will break plugins that implement a rest handler..., like wares, thrift, and memcached, until they are upgraded)
2012-06-28 02:00:47 +02:00
1ffd68f2de
Upgrade to netty 3.5.1
2012-06-28 00:51:37 +02:00
0aa0b9ef22
cleanups
2012-06-27 21:50:36 +02:00
d6bc17fee5
Partial update without script
...
Allow the use of "doc" as the update source when a script is not
specified. New fields are added, existing fields are overwritten, and
maps are merged recursively.
2012-06-27 21:29:22 +02:00
a4ad84b5e4
Enable validation of queries with has_child and script filters
2012-06-27 21:23:02 +02:00
dbeda1ab2b
Add missing serialization for error and explanation in validate query request
2012-06-27 21:23:01 +02:00
644dcb99cf
Setting index to no ends up using by default the "keyword" analyzer on it, which means one can't highlight on it when searching on _all, closes #2062 .
2012-06-27 21:16:03 +02:00
834c79597e
deb: Allow configuring max open files and max locked memory limits
2012-06-27 13:09:46 +02:00
d2bf87c89c
print the memory pools on GC logging
2012-06-27 12:05:53 +02:00
acc9783386
add 0.19.8
2012-06-26 19:59:40 +02:00
2b893fe1e5
Use bloom filter when flushing (applying deletes), closes #2058 .
2012-06-26 16:45:29 +02:00
2eee92726c
support also queue for setting queue size in fixed thread pool
2012-06-26 13:02:43 +02:00
7f77cfada0
Add support for zero queue size in the search thread pool
2012-06-26 13:01:18 +02:00
eb954a6157
better failure message when failing to load _uid field
2012-06-26 02:40:33 +02:00
12a644c89b
Stored Compression: failure to fetch document in certain cases (read failure, index compression works), closes #2055 .
2012-06-26 01:54:19 +02:00
6c74f99ed3
better generics in plugin module instance providers
2012-06-25 22:50:15 +02:00
c39a27a328
"index_analyzer" and "search_analyzer" should override "analyzer", closes #2052 .
2012-06-25 22:05:34 +02:00
11ca3ea0ec
deb package fails to start elasticsearch
...
The ability to use es.default prefix broke the deb package that uses it now.
2012-06-25 21:59:12 +02:00
5533fe2b88
don't double set settings with es.default. or elasticseearch.default.
...
though harmless, since we don't have any "default" component, and effectively has not effect on the settings used, it still puts settings that we don't really need
2012-06-25 21:31:18 +02:00
Shay Banon
Nicolas Lalevée
Matt Weber
Igor Motov
Petri Lehtinen