cassandra-model 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- data/.document +5 -0
- data/.gitignore +21 -0
- data/LICENSE +20 -0
- data/README.rdoc +56 -0
- data/Rakefile +90 -0
- data/VERSION +1 -0
- data/cassandra-model.gemspec +69 -0
- data/lib/cassandra-model.rb +27 -0
- data/lib/cassandra-model/base.rb +94 -0
- data/lib/cassandra-model/callbacks.rb +49 -0
- data/lib/cassandra-model/persistence.rb +112 -0
- data/lib/cassandra-model/types.rb +49 -0
- data/test/base_test.rb +45 -0
- data/test/callbacks_test.rb +43 -0
- data/test/cassandra_model_test.rb +81 -0
- data/test/config/cassandra.in.sh +47 -0
- data/test/config/log4j-tools.properties +27 -0
- data/test/config/log4j.properties +40 -0
- data/test/config/storage-conf.xml +368 -0
- data/test/test_helper.rb +10 -0
- metadata +112 -0
@@ -0,0 +1,49 @@
|
|
1
|
+
module CassandraModel
|
2
|
+
class StringType
|
3
|
+
def self.load(v)
|
4
|
+
v && v.to_s
|
5
|
+
end
|
6
|
+
end
|
7
|
+
|
8
|
+
class IntegerType
|
9
|
+
def self.load(v)
|
10
|
+
v && v.to_i
|
11
|
+
end
|
12
|
+
end
|
13
|
+
|
14
|
+
class FloatType
|
15
|
+
def self.load(v)
|
16
|
+
v && v.to_f
|
17
|
+
end
|
18
|
+
end
|
19
|
+
|
20
|
+
class DatetimeType
|
21
|
+
def self.dump(v)
|
22
|
+
v && v.strftime('%FT%T%z')
|
23
|
+
end
|
24
|
+
|
25
|
+
def self.load(v)
|
26
|
+
v && ::DateTime.strptime(v, '%FT%T%z')
|
27
|
+
end
|
28
|
+
end
|
29
|
+
|
30
|
+
class JsonType
|
31
|
+
def self.dump(v)
|
32
|
+
v && ::JSON.dump(v)
|
33
|
+
end
|
34
|
+
|
35
|
+
def self.load(v)
|
36
|
+
v && ::JSON.load(v)
|
37
|
+
end
|
38
|
+
end
|
39
|
+
|
40
|
+
class BooleanType
|
41
|
+
def self.dump(v)
|
42
|
+
v == '1'
|
43
|
+
end
|
44
|
+
|
45
|
+
def self.load(v)
|
46
|
+
v ? '1' : '0'
|
47
|
+
end
|
48
|
+
end
|
49
|
+
end
|
data/test/base_test.rb
ADDED
@@ -0,0 +1,45 @@
|
|
1
|
+
require File.expand_path(File.join(File.dirname(__FILE__), 'test_helper'))
|
2
|
+
|
3
|
+
class CassandraModelCallbacksTest < Test::Unit::TestCase
|
4
|
+
context "CassandraModel::Base" do
|
5
|
+
setup do
|
6
|
+
@klass = Class.new(CassandraModel::Base) do
|
7
|
+
key :name
|
8
|
+
column :age, :integer
|
9
|
+
column :dob, :datetime
|
10
|
+
column :note, :json
|
11
|
+
|
12
|
+
validate do
|
13
|
+
self.errors << "dob required" if dob.nil?
|
14
|
+
end
|
15
|
+
end
|
16
|
+
|
17
|
+
@klass.establish_connection 'cassandra-model'
|
18
|
+
end
|
19
|
+
|
20
|
+
should "connect to cassandra" do
|
21
|
+
assert_kind_of Cassandra, @klass.connection
|
22
|
+
end
|
23
|
+
|
24
|
+
should "store all defined columns" do
|
25
|
+
assert_equal({:age => :integer ,
|
26
|
+
:dob => :datetime,
|
27
|
+
:note => :json} , @klass.columns)
|
28
|
+
end
|
29
|
+
|
30
|
+
should "validate model by provided block" do
|
31
|
+
assert_kind_of Proc, @klass.validation
|
32
|
+
|
33
|
+
model = @klass.new()
|
34
|
+
assert !model.valid?
|
35
|
+
|
36
|
+
model = @klass.new(:name => "tl")
|
37
|
+
assert !model.valid?
|
38
|
+
|
39
|
+
model = @klass.new(:name => "tl", :dob => DateTime.now)
|
40
|
+
assert model.valid?
|
41
|
+
assert_equal "tl", model.key
|
42
|
+
assert_kind_of DateTime, model.dob
|
43
|
+
end
|
44
|
+
end
|
45
|
+
end
|
@@ -0,0 +1,43 @@
|
|
1
|
+
require File.expand_path(File.join(File.dirname(__FILE__), 'test_helper'))
|
2
|
+
|
3
|
+
class CassandraModelCallbacksTest < Test::Unit::TestCase
|
4
|
+
context "CassandraModel::Callbacks" do
|
5
|
+
setup do
|
6
|
+
@base = Class.new(Object) do
|
7
|
+
include CassandraModel::Callbacks
|
8
|
+
define_callbacks :foo
|
9
|
+
end
|
10
|
+
|
11
|
+
@klass = Class.new(@base) do
|
12
|
+
def bar; @n = [:bar]; end
|
13
|
+
|
14
|
+
def foo
|
15
|
+
run_callbacks(:foo) { @n << :foo }
|
16
|
+
end
|
17
|
+
|
18
|
+
def baz(v)
|
19
|
+
@n << :baz if v == [:bar, :foo]
|
20
|
+
end
|
21
|
+
|
22
|
+
def quux; @n << :quux; end
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
should "provide before and after callbacks for foo function" do
|
27
|
+
assert @klass.respond_to?(:define_callbacks)
|
28
|
+
assert @klass.respond_to?(:callbacks)
|
29
|
+
assert @klass.respond_to?(:before_foo)
|
30
|
+
assert @klass.respond_to?(:after_foo)
|
31
|
+
assert_equal Hash.new, @klass.callbacks
|
32
|
+
end
|
33
|
+
|
34
|
+
should "invoke callback functions when foo executed" do
|
35
|
+
@klass.send(:before_foo, :bar)
|
36
|
+
@klass.send(:after_foo, :baz, :quux)
|
37
|
+
assert_equal 2, @klass.callbacks.length
|
38
|
+
assert_equal [:bar], @klass.callbacks[:before_foo]
|
39
|
+
assert_equal [:baz, :quux], @klass.callbacks[:after_foo]
|
40
|
+
assert_equal [:bar, :foo, :baz, :quux], @klass.new.foo
|
41
|
+
end
|
42
|
+
end
|
43
|
+
end
|
@@ -0,0 +1,81 @@
|
|
1
|
+
|
2
|
+
require File.expand_path(File.join(File.dirname(__FILE__), 'test_helper'))
|
3
|
+
|
4
|
+
class User < CassandraModel::Base
|
5
|
+
column_family :Users
|
6
|
+
|
7
|
+
key :username
|
8
|
+
column :full_name
|
9
|
+
column :created_at, :datetime
|
10
|
+
|
11
|
+
write_consistency_level Cassandra::Consistency::ALL
|
12
|
+
|
13
|
+
before_save :set_default_time
|
14
|
+
|
15
|
+
validate do
|
16
|
+
errors << "full name required" if full_name.nil? || full_name.empty?
|
17
|
+
end
|
18
|
+
|
19
|
+
private
|
20
|
+
|
21
|
+
def set_default_time
|
22
|
+
self.created_at = Time.now
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
class CassandraModelTest < Test::Unit::TestCase
|
27
|
+
context "CassandraModel" do
|
28
|
+
setup do
|
29
|
+
@connection = CassandraModel::Base.establish_connection("CassandraModel")
|
30
|
+
@connection.clear_keyspace!
|
31
|
+
|
32
|
+
@user = User.create(:username => "tl", :full_name => "tien le")
|
33
|
+
end
|
34
|
+
|
35
|
+
should "be able to connect to Cassandra" do
|
36
|
+
assert_kind_of Cassandra, @connection
|
37
|
+
assert_equal "CassandraModel", @connection.keyspace
|
38
|
+
end
|
39
|
+
|
40
|
+
should "not create a new user when validation fails" do
|
41
|
+
user = User.create(:username => "tl")
|
42
|
+
assert !user.valid?
|
43
|
+
assert user.new_record?
|
44
|
+
|
45
|
+
user = User.new(:username => "tl").save
|
46
|
+
assert user.new_record?
|
47
|
+
assert_equal "full name required", user.errors.first
|
48
|
+
|
49
|
+
user = User.new(:full_name => "tl").save
|
50
|
+
assert_equal "key required", user.errors.first
|
51
|
+
end
|
52
|
+
|
53
|
+
should "create a new user when validation passed" do
|
54
|
+
assert !@user.new_record?
|
55
|
+
assert @user.eql?(User.get("tl"))
|
56
|
+
assert_equal @user, User.get("tl")
|
57
|
+
assert_equal "tien le", User.get("tl").full_name
|
58
|
+
|
59
|
+
user = User.new(:username => "abc", :full_name => "Foo")
|
60
|
+
user.save
|
61
|
+
assert_equal ["created_at", "full_name"], @connection.get(:Users, "abc").keys
|
62
|
+
end
|
63
|
+
|
64
|
+
should "destroy a record" do
|
65
|
+
@user.destroy
|
66
|
+
assert User.get("tl").nil?
|
67
|
+
assert_raise(CassandraModel::RecordNotFound) { User["tl"] }
|
68
|
+
end
|
69
|
+
|
70
|
+
should "return true if record exists and otherwise" do
|
71
|
+
assert User.exists?("tl")
|
72
|
+
assert !User.exists?("foo")
|
73
|
+
end
|
74
|
+
|
75
|
+
should "only take defined attributes" do
|
76
|
+
user = User.new(:username => "abc", :full_name => "Foo", :hachiko => 'dog')
|
77
|
+
user.save
|
78
|
+
assert_equal ["created_at", "full_name"], @connection.get(:Users, "abc").keys
|
79
|
+
end
|
80
|
+
end
|
81
|
+
end
|
@@ -0,0 +1,47 @@
|
|
1
|
+
# Licensed to the Apache Software Foundation (ASF) under one
|
2
|
+
# or more contributor license agreements. See the NOTICE file
|
3
|
+
# distributed with this work for additional information
|
4
|
+
# regarding copyright ownership. The ASF licenses this file
|
5
|
+
# to you under the Apache License, Version 2.0 (the
|
6
|
+
# "License"); you may not use this file except in compliance
|
7
|
+
# with the License. You may obtain a copy of the License at
|
8
|
+
#
|
9
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
10
|
+
#
|
11
|
+
# Unless required by applicable law or agreed to in writing, software
|
12
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
13
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
14
|
+
# See the License for the specific language governing permissions and
|
15
|
+
# limitations under the License.
|
16
|
+
|
17
|
+
# The directory where Cassandra's configs live (required)
|
18
|
+
CASSANDRA_CONF=$CASSANDRA_CONF
|
19
|
+
|
20
|
+
# This can be the path to a jar file, or a directory containing the
|
21
|
+
# compiled classes. NOTE: This isn't needed by the startup script,
|
22
|
+
# it's just used here in constructing the classpath.
|
23
|
+
cassandra_bin=$CASSANDRA_HOME/build/classes
|
24
|
+
|
25
|
+
# The java classpath (required)
|
26
|
+
CLASSPATH=$CASSANDRA_CONF:$CASSANDRA_BIN
|
27
|
+
|
28
|
+
for jar in $CASSANDRA_HOME/lib/*.jar $CASSANDRA_HOME/build/lib/jars/*.jar; do
|
29
|
+
CLASSPATH=$CLASSPATH:$jar
|
30
|
+
done
|
31
|
+
|
32
|
+
# Arguments to pass to the JVM
|
33
|
+
JVM_OPTS=" \
|
34
|
+
-ea \
|
35
|
+
-Xms128M \
|
36
|
+
-Xmx1G \
|
37
|
+
-XX:TargetSurvivorRatio=90 \
|
38
|
+
-XX:+AggressiveOpts \
|
39
|
+
-XX:+UseParNewGC \
|
40
|
+
-XX:+UseConcMarkSweepGC \
|
41
|
+
-XX:+CMSParallelRemarkEnabled \
|
42
|
+
-XX:+HeapDumpOnOutOfMemoryError \
|
43
|
+
-XX:SurvivorRatio=128 \
|
44
|
+
-XX:MaxTenuringThreshold=0 \
|
45
|
+
-Dcom.sun.management.jmxremote.port=8080 \
|
46
|
+
-Dcom.sun.management.jmxremote.ssl=false \
|
47
|
+
-Dcom.sun.management.jmxremote.authenticate=false"
|
@@ -0,0 +1,27 @@
|
|
1
|
+
# Licensed to the Apache Software Foundation (ASF) under one
|
2
|
+
# or more contributor license agreements. See the NOTICE file
|
3
|
+
# distributed with this work for additional information
|
4
|
+
# regarding copyright ownership. The ASF licenses this file
|
5
|
+
# to you under the Apache License, Version 2.0 (the
|
6
|
+
# "License"); you may not use this file except in compliance
|
7
|
+
# with the License. You may obtain a copy of the License at
|
8
|
+
#
|
9
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
10
|
+
#
|
11
|
+
# Unless required by applicable law or agreed to in writing, software
|
12
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
13
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
14
|
+
# See the License for the specific language governing permissions and
|
15
|
+
# limitations under the License.
|
16
|
+
|
17
|
+
# for production, you should probably set the root to INFO
|
18
|
+
# and the pattern to %c instead of %l. (%l is slower.)
|
19
|
+
|
20
|
+
# output messages into a rolling log file as well as stdout
|
21
|
+
log4j.rootLogger=WARN,stderr
|
22
|
+
|
23
|
+
# stderr
|
24
|
+
log4j.appender.stderr=org.apache.log4j.ConsoleAppender
|
25
|
+
log4j.appender.stderr.target=System.err
|
26
|
+
log4j.appender.stderr.layout=org.apache.log4j.PatternLayout
|
27
|
+
log4j.appender.stderr.layout.ConversionPattern=%5p %d{HH:mm:ss,SSS} %m%n
|
@@ -0,0 +1,40 @@
|
|
1
|
+
# Licensed to the Apache Software Foundation (ASF) under one
|
2
|
+
# or more contributor license agreements. See the NOTICE file
|
3
|
+
# distributed with this work for additional information
|
4
|
+
# regarding copyright ownership. The ASF licenses this file
|
5
|
+
# to you under the Apache License, Version 2.0 (the
|
6
|
+
# "License"); you may not use this file except in compliance
|
7
|
+
# with the License. You may obtain a copy of the License at
|
8
|
+
#
|
9
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
10
|
+
#
|
11
|
+
# Unless required by applicable law or agreed to in writing, software
|
12
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
13
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
14
|
+
# See the License for the specific language governing permissions and
|
15
|
+
# limitations under the License.
|
16
|
+
|
17
|
+
# for production, you should probably set the root to INFO
|
18
|
+
# and the pattern to %c instead of %l. (%l is slower.)
|
19
|
+
|
20
|
+
# output messages into a rolling log file as well as stdout
|
21
|
+
log4j.rootLogger=INFO,stdout,R
|
22
|
+
|
23
|
+
# stdout
|
24
|
+
log4j.appender.stdout=org.apache.log4j.ConsoleAppender
|
25
|
+
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
|
26
|
+
log4j.appender.stdout.layout.ConversionPattern=%5p %d{HH:mm:ss,SSS} %m%n
|
27
|
+
|
28
|
+
# rolling log file
|
29
|
+
log4j.appender.R=org.apache.log4j.RollingFileAppender
|
30
|
+
log4j.appender.file.maxFileSize=20MB
|
31
|
+
log4j.appender.file.maxBackupIndex=50
|
32
|
+
log4j.appender.R.layout=org.apache.log4j.PatternLayout
|
33
|
+
log4j.appender.R.layout.ConversionPattern=%5p [%t] %d{ISO8601} %F (line %L) %m%n
|
34
|
+
# Edit the next line to point to your logs directory
|
35
|
+
log4j.appender.R.File=data/logs/system.log
|
36
|
+
|
37
|
+
# Application logging options
|
38
|
+
#log4j.logger.com.facebook=DEBUG
|
39
|
+
#log4j.logger.com.facebook.infrastructure.gms=DEBUG
|
40
|
+
#log4j.logger.com.facebook.infrastructure.db=DEBUG
|
@@ -0,0 +1,368 @@
|
|
1
|
+
<!--
|
2
|
+
~ Licensed to the Apache Software Foundation (ASF) under one
|
3
|
+
~ or more contributor license agreements. See the NOTICE file
|
4
|
+
~ distributed with this work for additional information
|
5
|
+
~ regarding copyright ownership. The ASF licenses this file
|
6
|
+
~ to you under the Apache License, Version 2.0 (the
|
7
|
+
~ "License"); you may not use this file except in compliance
|
8
|
+
~ with the License. You may obtain a copy of the License at
|
9
|
+
~
|
10
|
+
~ http://www.apache.org/licenses/LICENSE-2.0
|
11
|
+
~
|
12
|
+
~ Unless required by applicable law or agreed to in writing,
|
13
|
+
~ software distributed under the License is distributed on an
|
14
|
+
~ "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
|
15
|
+
~ KIND, either express or implied. See the License for the
|
16
|
+
~ specific language governing permissions and limitations
|
17
|
+
~ under the License.
|
18
|
+
-->
|
19
|
+
<Storage>
|
20
|
+
<!--======================================================================-->
|
21
|
+
<!-- Basic Configurations -->
|
22
|
+
<!--======================================================================-->
|
23
|
+
|
24
|
+
<!--
|
25
|
+
~ The name of this cluster. This is mainly used to prevent machines in
|
26
|
+
~ one logical cluster from joining another.
|
27
|
+
-->
|
28
|
+
<ClusterName>Test Cluster</ClusterName>
|
29
|
+
|
30
|
+
<!--
|
31
|
+
~ Turn on to make new [non-seed] nodes automatically migrate the right data
|
32
|
+
~ to themselves. (If no InitialToken is specified, they will pick one
|
33
|
+
~ such that they will get half the range of the most-loaded node.)
|
34
|
+
~ If a node starts up without bootstrapping, it will mark itself bootstrapped
|
35
|
+
~ so that you can't subsequently accidently bootstrap a node with
|
36
|
+
~ data on it. (You can reset this by wiping your data and commitlog
|
37
|
+
~ directories.)
|
38
|
+
~
|
39
|
+
~ Off by default so that new clusters and upgraders from 0.4 don't
|
40
|
+
~ bootstrap immediately. You should turn this on when you start adding
|
41
|
+
~ new nodes to a cluster that already has data on it. (If you are upgrading
|
42
|
+
~ from 0.4, start your cluster with it off once before changing it to true.
|
43
|
+
~ Otherwise, no data will be lost but you will incur a lot of unnecessary
|
44
|
+
~ I/O before your cluster starts up.)
|
45
|
+
-->
|
46
|
+
<AutoBootstrap>false</AutoBootstrap>
|
47
|
+
|
48
|
+
<!--
|
49
|
+
~ Keyspaces and ColumnFamilies:
|
50
|
+
~ A ColumnFamily is the Cassandra concept closest to a relational
|
51
|
+
~ table. Keyspaces are separate groups of ColumnFamilies. Except in
|
52
|
+
~ very unusual circumstances you will have one Keyspace per application.
|
53
|
+
|
54
|
+
~ There is an implicit keyspace named 'system' for Cassandra internals.
|
55
|
+
-->
|
56
|
+
<Keyspaces>
|
57
|
+
<Keyspace Name="CassandraModel">
|
58
|
+
<!--
|
59
|
+
~ ColumnFamily definitions have one required attribute (Name)
|
60
|
+
~ and several optional ones.
|
61
|
+
~
|
62
|
+
~ The CompareWith attribute tells Cassandra how to sort the columns
|
63
|
+
~ for slicing operations. The default is BytesType, which is a
|
64
|
+
~ straightforward lexical comparison of the bytes in each column.
|
65
|
+
~ Other options are AsciiType, UTF8Type, LexicalUUIDType, TimeUUIDType,
|
66
|
+
~ and LongType. You can also specify the fully-qualified class
|
67
|
+
~ name to a class of your choice extending
|
68
|
+
~ org.apache.cassandra.db.marshal.AbstractType.
|
69
|
+
~
|
70
|
+
~ SuperColumns have a similar CompareSubcolumnsWith attribute.
|
71
|
+
~
|
72
|
+
~ BytesType: Simple sort by byte value. No validation is performed.
|
73
|
+
~ AsciiType: Like BytesType, but validates that the input can be
|
74
|
+
~ parsed as US-ASCII.
|
75
|
+
~ UTF8Type: A string encoded as UTF8
|
76
|
+
~ LongType: A 64bit long
|
77
|
+
~ LexicalUUIDType: A 128bit UUID, compared lexically (by byte value)
|
78
|
+
~ TimeUUIDType: a 128bit version 1 UUID, compared by timestamp
|
79
|
+
~
|
80
|
+
~ (To get the closest approximation to 0.3-style supercolumns, you
|
81
|
+
~ would use CompareWith=UTF8Type CompareSubcolumnsWith=LongType.)
|
82
|
+
~
|
83
|
+
~ An optional `Comment` attribute may be used to attach additional
|
84
|
+
~ human-readable information about the column family to its definition.
|
85
|
+
~
|
86
|
+
~ The optional KeysCached attribute specifies
|
87
|
+
~ the number of keys per sstable whose locations we keep in
|
88
|
+
~ memory in "mostly LRU" order. (JUST the key locations, NOT any
|
89
|
+
~ column values.) Specify a fraction (value less than 1), a percentage
|
90
|
+
~ (ending in a % sign) or an absolute number of keys to cache.
|
91
|
+
~ KeysCached defaults to 200000 keys.
|
92
|
+
~
|
93
|
+
~ The optional RowsCached attribute specifies the number of rows
|
94
|
+
~ whose entire contents we cache in memory. Do not use this on
|
95
|
+
~ ColumnFamilies with large rows, or ColumnFamilies with high write:read
|
96
|
+
~ ratios. Specify a fraction (value less than 1), a percentage (ending in
|
97
|
+
~ a % sign) or an absolute number of rows to cache.
|
98
|
+
~ RowsCached defaults to 0, i.e., row cache is off by default.
|
99
|
+
~
|
100
|
+
~ Remember, when using caches as a percentage, they WILL grow with
|
101
|
+
~ your data set!
|
102
|
+
-->
|
103
|
+
<ColumnFamily Name="Users" CompareWith="BytesType" KeysCached="100000" RowsCached="1000"/>
|
104
|
+
<ColumnFamily Name="Posts" CompareWith="BytesType" />
|
105
|
+
<ColumnFamily Name="Comments" CompareWith="TimeUUIDType" CompareSubcolumnsWith="BytesType" ColumnType="Super" />
|
106
|
+
|
107
|
+
<!--
|
108
|
+
~ Strategy: Setting this to the class that implements
|
109
|
+
~ IReplicaPlacementStrategy will change the way the node picker works.
|
110
|
+
~ Out of the box, Cassandra provides
|
111
|
+
~ org.apache.cassandra.locator.RackUnawareStrategy and
|
112
|
+
~ org.apache.cassandra.locator.RackAwareStrategy (place one replica in
|
113
|
+
~ a different datacenter, and the others on different racks in the same
|
114
|
+
~ one.)
|
115
|
+
-->
|
116
|
+
<ReplicaPlacementStrategy>org.apache.cassandra.locator.RackUnawareStrategy</ReplicaPlacementStrategy>
|
117
|
+
|
118
|
+
<!-- Number of replicas of the data -->
|
119
|
+
<ReplicationFactor>1</ReplicationFactor>
|
120
|
+
|
121
|
+
<!--
|
122
|
+
~ EndPointSnitch: Setting this to the class that implements
|
123
|
+
~ AbstractEndpointSnitch, which lets Cassandra know enough
|
124
|
+
~ about your network topology to route requests efficiently.
|
125
|
+
~ Out of the box, Cassandra provides org.apache.cassandra.locator.EndPointSnitch,
|
126
|
+
~ and PropertyFileEndPointSnitch is available in contrib/.
|
127
|
+
-->
|
128
|
+
<EndPointSnitch>org.apache.cassandra.locator.EndPointSnitch</EndPointSnitch>
|
129
|
+
|
130
|
+
</Keyspace>
|
131
|
+
</Keyspaces>
|
132
|
+
|
133
|
+
<!--
|
134
|
+
~ Authenticator: any IAuthenticator may be used, including your own as long
|
135
|
+
~ as it is on the classpath. Out of the box, Cassandra provides
|
136
|
+
~ org.apache.cassandra.auth.AllowAllAuthenticator and,
|
137
|
+
~ org.apache.cassandra.auth.SimpleAuthenticator
|
138
|
+
~ (SimpleAuthenticator uses access.properties and passwd.properties by
|
139
|
+
~ default).
|
140
|
+
~
|
141
|
+
~ If you don't specify an authenticator, AllowAllAuthenticator is used.
|
142
|
+
-->
|
143
|
+
<Authenticator>org.apache.cassandra.auth.AllowAllAuthenticator</Authenticator>
|
144
|
+
|
145
|
+
<!--
|
146
|
+
~ Partitioner: any IPartitioner may be used, including your own as long
|
147
|
+
~ as it is on the classpath. Out of the box, Cassandra provides
|
148
|
+
~ org.apache.cassandra.dht.RandomPartitioner,
|
149
|
+
~ org.apache.cassandra.dht.OrderPreservingPartitioner, and
|
150
|
+
~ org.apache.cassandra.dht.CollatingOrderPreservingPartitioner.
|
151
|
+
~ (CollatingOPP colates according to EN,US rules, not naive byte
|
152
|
+
~ ordering. Use this as an example if you need locale-aware collation.)
|
153
|
+
~ Range queries require using an order-preserving partitioner.
|
154
|
+
~
|
155
|
+
~ Achtung! Changing this parameter requires wiping your data
|
156
|
+
~ directories, since the partitioner can modify the sstable on-disk
|
157
|
+
~ format.
|
158
|
+
-->
|
159
|
+
<Partitioner>org.apache.cassandra.dht.RandomPartitioner</Partitioner>
|
160
|
+
|
161
|
+
<!--
|
162
|
+
~ If you are using an order-preserving partitioner and you know your key
|
163
|
+
~ distribution, you can specify the token for this node to use. (Keys
|
164
|
+
~ are sent to the node with the "closest" token, so distributing your
|
165
|
+
~ tokens equally along the key distribution space will spread keys
|
166
|
+
~ evenly across your cluster.) This setting is only checked the first
|
167
|
+
~ time a node is started.
|
168
|
+
|
169
|
+
~ This can also be useful with RandomPartitioner to force equal spacing
|
170
|
+
~ of tokens around the hash space, especially for clusters with a small
|
171
|
+
~ number of nodes.
|
172
|
+
-->
|
173
|
+
<InitialToken></InitialToken>
|
174
|
+
|
175
|
+
<!--
|
176
|
+
~ Directories: Specify where Cassandra should store different data on
|
177
|
+
~ disk. Keep the data disks and the CommitLog disks separate for best
|
178
|
+
~ performance
|
179
|
+
-->
|
180
|
+
<CommitLogDirectory>data/cassandra/commitlog</CommitLogDirectory>
|
181
|
+
<DataFileDirectories>
|
182
|
+
<DataFileDirectory>data/cassandra/data</DataFileDirectory>
|
183
|
+
</DataFileDirectories>
|
184
|
+
|
185
|
+
|
186
|
+
<!--
|
187
|
+
~ Addresses of hosts that are deemed contact points. Cassandra nodes
|
188
|
+
~ use this list of hosts to find each other and learn the topology of
|
189
|
+
~ the ring. You must change this if you are running multiple nodes!
|
190
|
+
-->
|
191
|
+
<Seeds>
|
192
|
+
<Seed>127.0.0.1</Seed>
|
193
|
+
</Seeds>
|
194
|
+
|
195
|
+
|
196
|
+
<!-- Miscellaneous -->
|
197
|
+
|
198
|
+
<!-- Time to wait for a reply from other nodes before failing the command -->
|
199
|
+
<RpcTimeoutInMillis>10000</RpcTimeoutInMillis>
|
200
|
+
<!-- Size to allow commitlog to grow to before creating a new segment -->
|
201
|
+
<CommitLogRotationThresholdInMB>128</CommitLogRotationThresholdInMB>
|
202
|
+
|
203
|
+
|
204
|
+
<!-- Local hosts and ports -->
|
205
|
+
|
206
|
+
<!--
|
207
|
+
~ Address to bind to and tell other nodes to connect to. You _must_
|
208
|
+
~ change this if you want multiple nodes to be able to communicate!
|
209
|
+
~
|
210
|
+
~ Leaving it blank leaves it up to InetAddress.getLocalHost(). This
|
211
|
+
~ will always do the Right Thing *if* the node is properly configured
|
212
|
+
~ (hostname, name resolution, etc), and the Right Thing is to use the
|
213
|
+
~ address associated with the hostname (it might not be).
|
214
|
+
-->
|
215
|
+
<ListenAddress>localhost</ListenAddress>
|
216
|
+
<!-- internal communications port -->
|
217
|
+
<StoragePort>7000</StoragePort>
|
218
|
+
|
219
|
+
<!--
|
220
|
+
~ The address to bind the Thrift RPC service to. Unlike ListenAddress
|
221
|
+
~ above, you *can* specify 0.0.0.0 here if you want Thrift to listen on
|
222
|
+
~ all interfaces.
|
223
|
+
~
|
224
|
+
~ Leaving this blank has the same effect it does for ListenAddress,
|
225
|
+
~ (i.e. it will be based on the configured hostname of the node).
|
226
|
+
-->
|
227
|
+
<ThriftAddress>localhost</ThriftAddress>
|
228
|
+
<!-- Thrift RPC port (the port clients connect to). -->
|
229
|
+
<ThriftPort>9160</ThriftPort>
|
230
|
+
<!--
|
231
|
+
~ Whether or not to use a framed transport for Thrift. If this option
|
232
|
+
~ is set to true then you must also use a framed transport on the
|
233
|
+
~ client-side, (framed and non-framed transports are not compatible).
|
234
|
+
-->
|
235
|
+
<ThriftFramedTransport>false</ThriftFramedTransport>
|
236
|
+
|
237
|
+
|
238
|
+
<!--======================================================================-->
|
239
|
+
<!-- Memory, Disk, and Performance -->
|
240
|
+
<!--======================================================================-->
|
241
|
+
|
242
|
+
<!--
|
243
|
+
~ Access mode. mmapped i/o is substantially faster, but only practical on
|
244
|
+
~ a 64bit machine (which notably does not include EC2 "small" instances)
|
245
|
+
~ or relatively small datasets. "auto", the safe choice, will enable
|
246
|
+
~ mmapping on a 64bit JVM. Other values are "mmap", "mmap_index_only"
|
247
|
+
~ (which may allow you to get part of the benefits of mmap on a 32bit
|
248
|
+
~ machine by mmapping only index files) and "standard".
|
249
|
+
~ (The buffer size settings that follow only apply to standard,
|
250
|
+
~ non-mmapped i/o.)
|
251
|
+
-->
|
252
|
+
<DiskAccessMode>auto</DiskAccessMode>
|
253
|
+
|
254
|
+
<!--
|
255
|
+
~ Size of compacted row above which to log a warning. (If compacted
|
256
|
+
~ rows do not fit in memory, Cassandra will crash. This is explained
|
257
|
+
~ in http://wiki.apache.org/cassandra/CassandraLimitations and is
|
258
|
+
~ scheduled to be fixed in 0.7.)
|
259
|
+
-->
|
260
|
+
<RowWarningThresholdInMB>512</RowWarningThresholdInMB>
|
261
|
+
|
262
|
+
<!--
|
263
|
+
~ Buffer size to use when performing contiguous column slices. Increase
|
264
|
+
~ this to the size of the column slices you typically perform.
|
265
|
+
~ (Name-based queries are performed with a buffer size of
|
266
|
+
~ ColumnIndexSizeInKB.)
|
267
|
+
-->
|
268
|
+
<SlicedBufferSizeInKB>64</SlicedBufferSizeInKB>
|
269
|
+
|
270
|
+
<!--
|
271
|
+
~ Buffer size to use when flushing memtables to disk. (Only one
|
272
|
+
~ memtable is ever flushed at a time.) Increase (decrease) the index
|
273
|
+
~ buffer size relative to the data buffer if you have few (many)
|
274
|
+
~ columns per key. Bigger is only better _if_ your memtables get large
|
275
|
+
~ enough to use the space. (Check in your data directory after your
|
276
|
+
~ app has been running long enough.) -->
|
277
|
+
<FlushDataBufferSizeInMB>32</FlushDataBufferSizeInMB>
|
278
|
+
<FlushIndexBufferSizeInMB>8</FlushIndexBufferSizeInMB>
|
279
|
+
|
280
|
+
<!--
|
281
|
+
~ Add column indexes to a row after its contents reach this size.
|
282
|
+
~ Increase if your column values are large, or if you have a very large
|
283
|
+
~ number of columns. The competing causes are, Cassandra has to
|
284
|
+
~ deserialize this much of the row to read a single column, so you want
|
285
|
+
~ it to be small - at least if you do many partial-row reads - but all
|
286
|
+
~ the index data is read for each access, so you don't want to generate
|
287
|
+
~ that wastefully either.
|
288
|
+
-->
|
289
|
+
<ColumnIndexSizeInKB>64</ColumnIndexSizeInKB>
|
290
|
+
|
291
|
+
<!--
|
292
|
+
~ Flush memtable after this much data has been inserted, including
|
293
|
+
~ overwritten data. There is one memtable per column family, and
|
294
|
+
~ this threshold is based solely on the amount of data stored, not
|
295
|
+
~ actual heap memory usage (there is some overhead in indexing the
|
296
|
+
~ columns).
|
297
|
+
-->
|
298
|
+
<MemtableThroughputInMB>64</MemtableThroughputInMB>
|
299
|
+
<!--
|
300
|
+
~ Throughput setting for Binary Memtables. Typically these are
|
301
|
+
~ used for bulk load so you want them to be larger.
|
302
|
+
-->
|
303
|
+
<BinaryMemtableThroughputInMB>256</BinaryMemtableThroughputInMB>
|
304
|
+
<!--
|
305
|
+
~ The maximum number of columns in millions to store in memory per
|
306
|
+
~ ColumnFamily before flushing to disk. This is also a per-memtable
|
307
|
+
~ setting. Use with MemtableThroughputInMB to tune memory usage.
|
308
|
+
-->
|
309
|
+
<MemtableOperationsInMillions>0.3</MemtableOperationsInMillions>
|
310
|
+
<!--
|
311
|
+
~ The maximum time to leave a dirty memtable unflushed.
|
312
|
+
~ (While any affected columnfamilies have unflushed data from a
|
313
|
+
~ commit log segment, that segment cannot be deleted.)
|
314
|
+
~ This needs to be large enough that it won't cause a flush storm
|
315
|
+
~ of all your memtables flushing at once because none has hit
|
316
|
+
~ the size or count thresholds yet. For production, a larger
|
317
|
+
~ value such as 1440 is recommended.
|
318
|
+
-->
|
319
|
+
<MemtableFlushAfterMinutes>60</MemtableFlushAfterMinutes>
|
320
|
+
|
321
|
+
<!--
|
322
|
+
~ Unlike most systems, in Cassandra writes are faster than reads, so
|
323
|
+
~ you can afford more of those in parallel. A good rule of thumb is 2
|
324
|
+
~ concurrent reads per processor core. Increase ConcurrentWrites to
|
325
|
+
~ the number of clients writing at once if you enable CommitLogSync +
|
326
|
+
~ CommitLogSyncDelay. -->
|
327
|
+
<ConcurrentReads>8</ConcurrentReads>
|
328
|
+
<ConcurrentWrites>32</ConcurrentWrites>
|
329
|
+
|
330
|
+
<!--
|
331
|
+
~ CommitLogSync may be either "periodic" or "batch." When in batch
|
332
|
+
~ mode, Cassandra won't ack writes until the commit log has been
|
333
|
+
~ fsynced to disk. It will wait up to CommitLogSyncBatchWindowInMS
|
334
|
+
~ milliseconds for other writes, before performing the sync.
|
335
|
+
|
336
|
+
~ This is less necessary in Cassandra than in traditional databases
|
337
|
+
~ since replication reduces the odds of losing data from a failure
|
338
|
+
~ after writing the log entry but before it actually reaches the disk.
|
339
|
+
~ So the other option is "periodic," where writes may be acked immediately
|
340
|
+
~ and the CommitLog is simply synced every CommitLogSyncPeriodInMS
|
341
|
+
~ milliseconds.
|
342
|
+
-->
|
343
|
+
<CommitLogSync>periodic</CommitLogSync>
|
344
|
+
<!--
|
345
|
+
~ Interval at which to perform syncs of the CommitLog in periodic mode.
|
346
|
+
~ Usually the default of 10000ms is fine; increase it if your i/o
|
347
|
+
~ load is such that syncs are taking excessively long times.
|
348
|
+
-->
|
349
|
+
<CommitLogSyncPeriodInMS>10000</CommitLogSyncPeriodInMS>
|
350
|
+
<!--
|
351
|
+
~ Delay (in milliseconds) during which additional commit log entries
|
352
|
+
~ may be written before fsync in batch mode. This will increase
|
353
|
+
~ latency slightly, but can vastly improve throughput where there are
|
354
|
+
~ many writers. Set to zero to disable (each entry will be synced
|
355
|
+
~ individually). Reasonable values range from a minimal 0.1 to 10 or
|
356
|
+
~ even more if throughput matters more than latency.
|
357
|
+
-->
|
358
|
+
<!-- <CommitLogSyncBatchWindowInMS>1</CommitLogSyncBatchWindowInMS> -->
|
359
|
+
|
360
|
+
<!--
|
361
|
+
~ Time to wait before garbage-collection deletion markers. Set this to
|
362
|
+
~ a large enough value that you are confident that the deletion marker
|
363
|
+
~ will be propagated to all replicas by the time this many seconds has
|
364
|
+
~ elapsed, even in the face of hardware failures. The default value is
|
365
|
+
~ ten days.
|
366
|
+
-->
|
367
|
+
<GCGraceSeconds>864000</GCGraceSeconds>
|
368
|
+
</Storage>
|