ambari-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mehmet Zahid Yüzügüldü <>
Subject about apache ambari workflow
Date Mon, 24 Aug 2015 12:59:56 GMT
Hi guys

i make lots of  research about ambari.

i am using hadoop cluster about 2 years. HBase, Yarn Ozie, Hue and Solr.

i am trying to add some dummy skills into ambari to learn the main

my object is to add ambari functionality that disable selinux from nodes
with the help of ambari-server and ambari-agent usage.

java spring was used to develope ambari server, i tracked some api calls
from ambari-web.

i got the url "http://node1:8080/api/v1/requests
<>"  with request body:

  "RequestInfo": {
    "action": "check_host",
    "context": "Check host",
    "parameters": {
      "check_execute_list": "host_resolution_check",
      "jdk_location": "http://node1:8080/resources/",
      "threshold": "20",
      "hosts": "node1 <>"
  "Requests/resource_filters": [
      "hosts": "node1 <>"

i found js file related with this call in

it is creating the request and calls rest api as i said :

but i couldn't find the mapping of this rest api. i mean what service
method was fired with this rest api, i couldn't find.

after i found this, i will insert new record into db with information to
disable selinux, like script type, script name to be run etc.

   "hostname":"node1 <>",







so when *ambari-agent heardbeat*, it will get that record and will run
script to disable selinux. am i wrong?

i wonder if you could you help me to do that.

i appreciate that.

Mehmet Zahid Yüzügüldü

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message