spark_toolkit 0.1.0-java → 0.1.1-java
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/README.md +1 -1
- data/docs/Spark.md +1 -0
- data/lib/spark_toolkit/hadoop/conf/configuration.rb +7 -0
- data/lib/spark_toolkit/spark/client.rb +12 -3
- data/lib/spark_toolkit/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 56d3aadfc0f28d592c7051ee0fe9427df5f73383
|
4
|
+
data.tar.gz: 0fa58599de96d34d0c1fcf613e288073867ce6b8
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 20f30636b201ede56541e852fce3828211e4ee3aad31f83155afd8b58168a143d6c8c8bac913101be49790aa4acc601c55f2457039c722da9d0f2e666db72b4b
|
7
|
+
data.tar.gz: 053cf2ee78cd8bf2707aaa0c255ce21cb0866bf56b43b84056a262b8d8b1b2f9f570f6869dbad6874f6997064388a48bc81e2d7932163b4dd3801077572901e4
|
data/README.md
CHANGED
data/docs/Spark.md
CHANGED
@@ -22,6 +22,7 @@ args = ["--class", "org.apache.spark.deploy.PythonRunner",
|
|
22
22
|
"--arg", 2]
|
23
23
|
spark_conf.yarn_deploy_mode(:cluster) # or :client
|
24
24
|
spark_client.is_python_job(true)
|
25
|
+
spark_client.avtive_kerberos # If you want to submit job to secure cluster
|
25
26
|
# Submit your job to YARN and get its app_id for query
|
26
27
|
yarn_app_id = spark_client.yarn_submit(args)
|
27
28
|
# Or run as client, print all output into console
|
@@ -3,6 +3,7 @@ module SparkToolkit
|
|
3
3
|
Configuration = Java::OrgApacheHadoopConf::Configuration
|
4
4
|
class Configuration
|
5
5
|
java_import org.apache.hadoop.fs.Path
|
6
|
+
java_import org.apache.hadoop.security.UserGroupInformation
|
6
7
|
|
7
8
|
alias_method :initialise, :initialize
|
8
9
|
def initialize(opts={})
|
@@ -21,6 +22,12 @@ module SparkToolkit
|
|
21
22
|
add_resource_java(Path.new(f))
|
22
23
|
end
|
23
24
|
|
25
|
+
def krb_login(principle, keytab)
|
26
|
+
set('hadoop.security.authentication', 'kerberos')
|
27
|
+
UserGroupInformation.set_configuration(self)
|
28
|
+
UserGroupInformation.login_user_from_keytab(principle, keytab)
|
29
|
+
end
|
30
|
+
|
24
31
|
def []=(k, v)
|
25
32
|
set(k, v)
|
26
33
|
end
|
@@ -69,9 +69,18 @@ module SparkToolkit
|
|
69
69
|
@sconf.set("spark.hadoop.hadoop.security.authentication", "kerberos")
|
70
70
|
@sconf.set("spark.hadoop.hadoop.security.authorization", "true")
|
71
71
|
|
72
|
+
UserGroupInformation.get_login_user.check_tgt_and_relogin_from_keytab
|
72
73
|
UserGroupInformation.set_configuration(SparkHadoopUtil.get.newConfiguration(@sconf))
|
73
|
-
credentials = UserGroupInformation.
|
74
|
-
SparkHadoopUtil.get.
|
74
|
+
credentials = UserGroupInformation.get_login_user.get_credentials
|
75
|
+
SparkHadoopUtil.get.add_current_user_credentials(credentials)
|
76
|
+
end
|
77
|
+
|
78
|
+
def executor_cores n
|
79
|
+
@sconf.set_property('spark.executor.cores', n.to_s)
|
80
|
+
end
|
81
|
+
|
82
|
+
def num_executors n
|
83
|
+
@sconf.set_property('spark.executor.instances', n.to_s)
|
75
84
|
end
|
76
85
|
|
77
86
|
private
|
@@ -83,7 +92,7 @@ module SparkToolkit
|
|
83
92
|
@sconf.set('spark.submit.deployMode', 'cluster')
|
84
93
|
end
|
85
94
|
|
86
|
-
java.lang.System.
|
95
|
+
java.lang.System.set_property("SPARK_YARN_MODE", "true")
|
87
96
|
end
|
88
97
|
end
|
89
98
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: spark_toolkit
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.1
|
5
5
|
platform: java
|
6
6
|
authors:
|
7
7
|
- Yuli Mo
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2017-02-
|
11
|
+
date: 2017-02-08 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|