[
https://issues.apache.org/jira/browse/HADOOP-6944?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12912792#action_12912792
]
Konstantin Boudnik commented on HADOOP-6944:
--------------------------------------------
I suggest to make the following changes:
- change the signature of {{getHadoopMultiUsersList}} to
{{public ProxyUserDefinitions getHadoopMultiUsersList() throws IOException}}
and then
{noformat}
public abstract class ProxyUserDefinitions {
class GroupsAndHost {
private List<String> groups;
private List<String> hosts;
public List<String> getGroups() {
return groups;
}
public void setGroups(List<String> groups) {
this.groups = groups;
}
public List<String> getHosts() {
return hosts;
}
public void setHosts(List<String> hosts) {
this.hosts = hosts;
}
}
protected Map<String, GroupsAndHost> proxyUsers;
protected ProxyUserDefinitions () {
proxyUsers = new HashMap<String, GroupsAndHost>();
}
public void addProxyUser (String userName, GroupsAndHost definitions) {
proxyUsers.put(userName, definitions);
}
public GroupsAndHost getProxyUser (String userName) {
return proxyUsers.get(userName);
}
/**
* The implementation of this method has to be provided by a child of the
class
* @param filePath
* @return
* @throws IOException
*/
public abstract boolean writeToFile(URI filePath) throws IOException;
}{noformat}
so you will be able to store information in this form {{username:
group[,group]*}} as well as the hostnames from where proxy users are allowed to
connect. The purpose of this new method will be to extract only proxy-users
related information from a daemon configuration and present it in the form a
data container (above).
The data container provides the method {{writeToFile}} which has to be
overwritten by an implementer in order to provide needed logic of how the
content of object is stored into a file. Thus, you can inherit the class in
your testcase (or whatever) and create writeToFile implementation you need
(basically, the second method {{getProxyUsersFile}} of your patch with some
small modificaitons) for gridmix without adding unrelated functionality to
Herriot base classes.
Makes sense?
> [Herriot] Implement a functionality for building proxy users file and proxy
> users list.
> ---------------------------------------------------------------------------------------
>
> Key: HADOOP-6944
> URL: https://issues.apache.org/jira/browse/HADOOP-6944
> Project: Hadoop Common
> Issue Type: Task
> Components: test
> Reporter: Vinay Kumar Thota
> Assignee: Vinay Kumar Thota
> Attachments: HADOOP-6944.patch, HADOOP-6944.patch
>
>
> Gridmix should require a proxy user's file for impersonating various jobs.
> So, implement couple of methods for getting the proxy users list and a proxy
> users file (it's a combination of proxy users and groups) based on cluster
> configuration.
> The proxy users list should require for map reduce jobs and proxy users file
> should require for gridmix jobs.
> The following are methods signature,
> public String getProxyUsersFile() - getting the proxy users file based on
> cluster configuration information.
> public String buildProxyUsersFile() - Build the proxy users file using the
> cluster configuration.
> public ArraryList<String> getHadoopMultiUsersList() - get the list of proxy
> users list based on cluster configuration.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
