forked from linkedin/dr-elephant
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathJobTypeConf.xml
57 lines (56 loc) · 2.06 KB
/
JobTypeConf.xml
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
<?xml version="1.0" encoding="UTF-8"?>
<!--
This file defines MapReduce jobtype identified by Dr. Elephant
Dr. will match jobtype by looking at particular MR job conf property and value
defined for that job type. For example, if a MR job has 'pig.script' in its configuration,
it will be classified as Pig jobtype; if a MR job has configuration 'mapred.reducer.class'
and value 'HadoopStoreBuilderReducer', it will be a voldemort job.
For each jobtype defined in this file, it is required to have a type name <name>, a conf
name <conf>, and an optional conf value <value>. The optional <value> requires a regex
expression with '.*'(any char multiple times) by default.
<isDefault/> when present is indicating the current JobType is the default choice for an application type. One
application type should have one and only one default choice.
Type matching is top-down - If a job matches multiple types in this file, it will be First
Defined Comes First. If no jobtype in this file matches the job, it will have default jobtype,
which is HadoopJava.
-->
<jobTypes>
<jobType>
<name>Spark</name>
<applicationtype>spark</applicationtype>
<conf>spark.app.id</conf>
<isDefault/>
</jobType>
<jobType>
<name>Pig</name>
<applicationtype>mapreduce</applicationtype>
<conf>pig.script</conf>
</jobType>
<jobType>
<name>Hive</name>
<applicationtype>mapreduce</applicationtype>
<conf>hive.mapred.mode</conf>
</jobType>
<jobType>
<name>Cascading</name>
<applicationtype>mapreduce</applicationtype>
<conf>cascading.app.frameworks</conf>
</jobType>
<jobType>
<name>Voldemort</name>
<applicationtype>mapreduce</applicationtype>
<conf>mapred.reducer.class</conf>
<value>voldemort.store.readonly.mr.*</value>
</jobType>
<jobType>
<name>Kafka</name>
<applicationtype>mapreduce</applicationtype>
<conf>kafka.url</conf>
</jobType>
<jobType>
<name>HadoopJava</name>
<applicationtype>mapreduce</applicationtype>
<conf>mapred.child.java.opts</conf>
<isDefault/>
</jobType>
</jobTypes>