2 # Copyright 2013-2014, whitestar
4 # Licensed under the Apache License, Version 2.0 (the "License");
5 # you may not use this file except in compliance with the License.
6 # You may obtain a copy of the License at
8 # http://www.apache.org/licenses/LICENSE-2.0
10 # Unless required by applicable law or agreed to in writing, software
11 # distributed under the License is distributed on an "AS IS" BASIS,
12 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 # See the License for the specific language governing permissions and
14 # limitations under the License.
17 $LOAD_PATH.push("#{File.dirname(__FILE__)}/_lib")
18 #${chef-repo}/roles/_lib/grid.rb
21 cluster_name = 'ptolemy'
22 cluster_sym = cluster_name.to_sym
24 name "hadoop-#{cluster_name}"
25 description "#{cluster_name.capitalize} Hadoop cluster node"
30 #'role[nameservice-client-grid]',
36 grid_vol_root = '/grid/vol'
39 'this.cluster.name' => cluster_name,
40 'this.domain' => Grid::DOMAIN,
41 'this.namenode.fqdn' => '${this.cluster.name}-nn.${this.domain}',
42 'fs.checkpoint.dir' => "#{grid_vol_root}/0/var/lib/${user.name}/checkpoint",
43 #'fs.checkpoint.dir' => "#{grid_vol_root}/0/var/lib/${user.name}/checkpoint,/export/home/${user.name}/var/lib/checkpoint",
45 'this.secondary.namenode.fqdn' => '${this.cluster.name}-cn.${this.domain}',
46 'dfs.name.dir' => "#{grid_vol_root}/0/var/lib/${user.name}/name",
47 #'dfs.name.dir' => "#{grid_vol_root}/0/var/lib/${user.name}/name,/export/home/${user.name}/var/lib/name",
48 'dfs.replication' => '2',
49 #'dfs.replication' => '3', # default, recommended.
51 'this.jobtracker.fqdn' => '${this.cluster.name}-jt.${this.domain}',
52 'this.jobhistory.fqdn' => '${this.cluster.name}-jt.${this.domain}',
53 #'this.jobhistory.fqdn' => '${this.cluster.name}-jh.${this.domain}',
55 # e.g. 'core-site.xml' => {'k1' => 'v1', 'k2' => 'v2'},
57 #'HADOOP_HEAPSIZE' => '2000', # default: 1000 MB
58 #'HADOOP_NAMENODE_INIT_HEAPSIZE' => '2000' # default: 1000 MB (ver. 2.x)
59 #'HADOOP_NAMENODE_OPTS' => '',
60 #'HADOOP_SECONDARYNAMENODE_OPTS' => '',
61 #'HADOOP_DATANODE_OPTS' => '',
62 #'HADOOP_BALANCER_OPTS' => '',
63 #'HADOOP_JOBTRACKER_OPTS' => '',
65 'yarn-env.sh' => { # ver. 2.x
66 #'YARN_HEAPSIZE' => '2000' # default: 1000 MB
67 #'YARN_RESOURCEMANAGER_OPTS' => '',
68 #'YARN_NODEMANAGER_OPTS' => '',
70 'mapred-env.sh' => { # ver. 2.x
71 #'HADOOP_JOB_HISTORYSERVER_HEAPSIZE' => '2000', # default: 1000 MB
72 #'HADOOP_JOB_HISTORYSERVER_OPTS' => '',
77 'dfs.block.size' => '134217728', # default: 67108864 (bytes)
78 #'dfs.blocksize' => '134217728', # default: 134217728 (bytes) (ver. 2.x)
79 #'dfs.datanode.du.reserved' => '0', # default: 0 (byte)
81 #'dfs.support.append' => 'true',
82 #'dfs.datanode.max.xcievers' => '4096', # default: 4096 (ver. 1.1 or later), 256 (older ver.)
83 #'dfs.datanode.max.transfer.threads' => '4096', # default: 4096 (ver. 2.x)
85 'yarn-site.xml' => { # ver. 2.x
86 #'yarn.nodemanager.resource.cpu-vcores' => '8', # default: 8
87 #'yarn.nodemanager.resource.memory-mb' => '8192', # default: 8192
89 'mapred-site.xml' => {
90 # map tasks:reduce tasks = 4:3 or 2:1 or ...
91 'mapred.tasktracker.map.tasks.maximum' => '4', # default: 2
92 'mapred.tasktracker.reduce.tasks.maximum' => '2', # default: 2
93 #'mapred.child.java.opts' => '-Xmx1000m', # default: -Xmx200m
94 'mapred.reduce.slowstart.completed.maps' => '0.80', # default: 0.05
95 #'mapreduce.job.reduce.slowstart.completedmaps' => '0.80', # default: 0.05 (ver. 2.x)
96 #'tasktracker.http.threads' => '80', # default: 40
97 #'mapreduce.tasktracker.http.threads' => '80', # default: 40 (ver. 2.x)
98 #'mapred.reduce.parallel.copies' => SQRT(nodes), # default: 5
99 #'mapreduce.reduce.shuffle.parallelcopies' => SQRT(nodes), # default: 5 (ver. 2.x)
100 'mapred.submit.replication' => '2', # default: 10
101 #'mapreduce.client.submit.file.replication' => '3', # default: 10 (ver. 2.x)
103 'capacity-scheduler.xml' => {}
107 #default_attributes()
111 'vol_root' => grid_vol_root,
112 'max_vol_nums' => '4'
114 'hadoop' => hadoop_attributes