forked from sgroschupf/katta
-
Notifications
You must be signed in to change notification settings - Fork 0
/
CHANGES.txt
188 lines (175 loc) · 11.2 KB
/
CHANGES.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
Katta Change Log
katta 0.7.0
fix KATTA-226, configurable "track-doc-scores-on-search-with-field-sort"
fix KATTA-221, make LuceneServer.SearcherHandle protected
fix KATTA-216, fixed issues that could occur when a Node is disconnected/reconnected during initialization
fix KATTA-206, fix ConcurrentModificationException in One2ManyListMap
fix KATTA-197, switch to node selection policy that shuffles the order before connecting to nodes
fix KATTA-201, Added new HostAwareDistributionPolicy for use when running multiple nodes on the same host
fix KATTA-209, ConcurrentModificationException in Client _indexToShards HashMap
fix KATTA-215, Field sort is not deterministic if two Hits have the same field values
fix KATTA-214, Relevance ordering inconsistent with Lucene doc ordering
fix KATTA-211, Hits sort not deterministic
fix KATTA-194, Upgrade to lucene 3.5.0 (from 3.0.3)
fix KATTA-180, better shard-removal handling in client
fix KATTA-198, expose missed shards for query to Hits object
fix KATTA-205, close IndexSearcher only after all running searches completed
fix KATTA-204, Allow KattaMiniCluster to be used without depending on config files on the classpath
fix KATTA-203, Lucene search fails if a deployed index has zero documents
fix KATTA-202, Client exceptions don't always get logged if they are caught after the result set is closed
fix KATTA-200, fix ShardManagerTest on fast computers
fix KATTA-199, drop katta gui groovy project
fix KATTA-195, simplify Katta build
katta 0.6.5
fix KATTA-226, configurable "track-doc-scores-on-search-with-field-sort"
fix KATTA-196, LuceneServer leaks memory when using query filters
fix KATTA-193, implement query filters
katta 0.6.4
fix KATTA-186, make message about document and term statistics more readable
fix KATTA-189, improve multithreaded shard search by using ExecutorCompletionService
fix KATTA-190, make inner classes of LuceneServer protected
fix KATTA-187, add additional parameters to the Katta client to change the output format for easier scripting and parsing
fix KATTA-192, make master failover more robust
fix KATTA-184, upgrade to zookeeper 3.3.2 (from 3.3.1)
fix KATTA-183, zkclient can get unresponsive through OOM
fix KATTA-182, fix memory leak in client - ZooKeeper$ZkWatchManager.existWatches can grow huge
fix KATTA-181, don't remove node-shard-mapping from client if proxy fails one time
fix KATTA-179, undeployment of index can lead to NPE in BooleanQuery - improve exception message
fix KATTA-178, undeploying indices can leave empty shard-to-node pathes
fix KATTA-177, IndexDeployFuture not safe for quick undeployments
fix KATTA-80, add jar-test target to build which builds a jar of test classes
fix KATTA-175, make lucene-server thread pool parameters configurable
fix KATTA-171, use timeout from Client in the LuceneServer as well
fix KATTA-174, allow configuration of content-server through katta.node.properties
fix KATTA-173, upgrade to hadoop-0.20.2 (from 0.20.1)
fix KATTA-172, upgrade to lucene-3.0.3 (from 3.0.0)
katta 0.6.3
fix KATTA-165, fix IndexOutOfBoundsException when adding index with enabled throttling
fix KATTA-161, fix deadlock situation when node gets disconnected from zk and so becomes unresponsive
fix KATTA-162, Allow LuceneClient to be extended more easily
fix KATTA-156, allow shard selection by regular expression
fix KATTA-163, don't exit node/master operation thread in case an unexpected exception occurs
fix KATTA-160, Stop trying to rebalance/replicate an index when the index could not be found in the file system any more
fix KATTA-134, cobertura.jar version mismatch when compiling /extras/indexing
fix KATTA-153, LuceneServer loads all fields from index, even if only fewer are requested
fix KATTA-157, set timeout on LuceneClient
fix KATTA-154, HitsMapWritable readFields does not add hits optimally
fix KATTA-152, modify LuceneServer for easier sub-classing
fix KATTA-159, LuceneServerTest does not compile
fix KATTA-150, port parameter for startNode command
fix KATTA-149, remove unnecessary synchronization from LuceneServer
fix KATTA-148, fix debug statement which prevent katta from running with debug log level
fix KATTA-147, upgrade to zookeeper 3.3.1
katta 0.6.2
fix KATTA-146, fix ConcurrentModificationException when multiple LuceneClient objects are created simultaneously
fix KATTA-145, fix NullPointerException when a LuceneClient is created and an index is reomoved concurrently
fix KATTA-142, fix memory leak in client usage when adding and removing indices
fix KATTA-136, fix NPE in client on remove index event
fix KATTA-141, fix ec2 scripts
fix KATTA-128, set node RPC handler count from 1 to 25 and make it configurable
fix KATTA-130, fix unzipping, when parent dirs do not exists
fix KATTA-126, some improvements in use of java generics
fix KATTA-140, fix search when number of documents exceeds Integer.MAX_VALUE
fix KATTA-135, fix NPE in AbstractIndexOperation.addRunningDeployments
fix KATTA-125, fix ConcurrentModificationException in net.sf.katta.protocol.InteractionProtocol
katta 0.6.1
fix KATTA-122, removeIndex leaks file descriptors
fix KATTA-124, fix imbalanced shard distribution
katta 0.6.0
fix KATTA-120, fix listIndices for wrong file pathes
fix KATTA-117, add command line option to print stacktrace on error
fix KATTA-116, fix distribution of shards does not take currently deploying shards into account
fix KATTA-107, fix katta execution on cygwin
fix KATTA-112, ship build.xml in core distribution
fix KATTA-110, use a released 0.1 version of zkclient instead of the snapshot
katta 0.6.rc1
fix KATTA-109, split katta distribution into katta and katta.gui
fix KATTA-17, load test for katta
fix KATTA-96, upgrade mechanism for katta
fix KATTA-104, upgrade to zookeeper 3.2.2
fix KATTA-105, throttle shard deployment
fix KATTA-103, upgrade to lucene 3.0
fix KATTA-82, Katta need to be monitor able
fix KATTA-101, refactore INodeManaged implementation into sub-packages
fix KATTA-97, gracefull shutdown of JmxMonitor
fix KATTA-102, node failover in Client is not safe for multithreaded use
fix KATTA-95, IndexDeployFuture.joinDeployment() seems to hang from time to time
fix KATTA-43, Katta does not recover well from expired sessions
fix KATTA-100, ivy setup does not work for extras/indexing module
fix KATTA-93, hits are (re-)sorted completely on client side
fix KATTA-78, Add basic Lucene Sort capabilities
fix KATTA-76, fix listIndexes in case unreachable indices are deployed
fix KATTA-81, NodeInteraction: configurable max-try-count
fix KATTA-54, Generalize Katta so that Lucene is one use case. Add MapFile implementation
fix KATTA-73, CDPATH environment variable causes bin scripts to fail.
fix KATTA-74, upgrading dependencies for extras/indexer
fix KATTA-55, adding /extras to the distribution
fix KATTA-75, tarbomb: Katta 0.5.1 release tarball expands in place
fix KATTA-80, Configuration loading from files from Jason Rutherglen
fix KATTA-56, fixing a comment refering to hadoop instead to katta
fix KATTA-86, sources should be in jar as well.
fix KATTA-77, removing indexer and merging code, since it is obsolete. Indexer sample code can be find in the extras/indexer folder. Instead of merging indexes we recommend to reindex the complete data set.
fix KATTA-84, adding decompressing during download without storing it on the hdd - from Jason Venner
fix KATTA-84, adding a LowestShardCountDistributionPolicy from Jason Venner
fix KATTA-87, adding a first version of the katta.gui
fix KATTA-85, upgrade to hadoop 0.20.1 jars
fix KATTA-57, Permissions on scripts are now set properly with execute flag set in .tar.gz file
fix KATTA-48, Index name is now included in IndexMetaData
fix KATTA-44, Cleaned up ordering of Hits in KattaHitQueue
fix KATTA-42, Moved to Zookeeper-3.2.1.
remove maven repos from ivy configuration and add last missing libraries into lib folder
fix KATTA-72, Give better error message when searching an index that doesn't exist.
fix KATTA-67, Default namespace is not created when using external Katta
fix KATTA-66, Update jets3t jar from version 0.5.0 to 0.6.1
fix KATTA-71, Katta hangs when deploying index from s3 since HADOOP-4422
fix KATTA-70, When using an external Zookeeper master and secondary master don't terminate straight away anymore.
fix KATTA-63, Use java found on the path, if JAVA_HOME is not set
fixing a missleading path in the katta-evn.sh
fix KATTA-14, Katta should be able to use an external zookeeper cluster.
fix KATTA-53, fix spelling errors and other small issues in code, contributed by Ted Dunning.
fix KATTA-47, anaylzer is not required anymore, contributed by Vivek Magotra.
fix KATTA-52, ZKClient.reconnect() should only be called on KeeperState.Expired events
fix KATTA-51, master hangs when reconnecting to zk during deployment of a index.
fix KATTA-46, constructor for client and DeployClient exposing zkClient configuration values, so no ZKConfiguration object is required.
fix KATTA-45, we might ignored hits with a higher score that are in later shards.
merged katta 0.5.1 into 0.6-dev
fix KATTA-41, SecondaryMaster cannot take over when firstMaster failed.
fix KATTA-39, logical bug in calculating the DF
fix KATTA-37 (blocker), undeploy shard also deleted other shards
fix KATTA-38, endless loop in finding free rcp port
KATTA-17, Added first cut load testing code.
fix KATTA-36, bin/katta showStructure not throwing an exception anymore.
fix KATTA-35, Added missing target descriptions to build.xml.
fix KATTA-31, dont rsync log or zookeeper folders
katta 0.5
fix KATTA-23, Parallize result detail retrieval
fix KATTA-2, use the correct analyzer for queries. I used the opportunity here to depreciate IQuery and Query and switched it to lucene Query.
It is now possible to use any kind of lucene query created programatically. If you using a string you can use the lucene QueryParser.
The depreciate still uses the hardcoded KeywordAnalyzer.
fix NPE in Node zookeeper reconnect in case the shardfolder was empty.
fix KATTA-4, shards are now searched in parallel with a thread pool, we still need to merge the results together though.
fix KATTA-27, Query parser not thread-safe.
minor refactoring to make Node better testable without zookeeper.
fix KATTA-10, add port to used shard folder so it is possible to start multiple nodes on one server
fix KATTA-25, upgrading to zookeeper 3.1.1
fix KATTA-20, Hits is now thread safe, because it is used by multiple searchers that are running concurrently to collect their results
DocumentFrequenceWritable is now thread safe doc and term frequencies are always computed correctly
fix KATTA-24, update to lucene 2.4.1
fix KATTA-21
fix KATTA-22
SampleIndex generator uses hostname plus uuid for index name now.
Restructure build to support multi sub projects.
Adding coverage reports
switching to ant and ivy for build
upgrading to hadoop 0.19.
adding ec2 support
Release 0.4.0 - 2008-12-08
upgrade hadoop to 0.18.1
store full stacktrace of index-failures
add merging of indexes
respect node status 'STARTING' in SAFE_MODE
fix bug where node jvm exits on startup on an deployment error
check for indexes in deploying and replicating state on master startup
upgrade zookeeper version from 2.2.0 to 2.2.1
Release 0.1.0 - 2008-10-17
1. The first release of Katta.