[ 
https://issues.apache.org/jira/browse/MAPREDUCE-3532?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13177652#comment-13177652
 ] 

Bhallamudi Venkata Siva Kamesh commented on MAPREDUCE-3532:
-----------------------------------------------------------

When we set 0 as port for *NM_WEBAPP_ADDRESS*,  *this.httpPort* will be 
assigned the same value. While sending registeration requestion to the RM, NM 
sends the same http port as part of its request. In this case http port happens 
0.

{code:title=NodeStatusUpdaterImpl.java|borderStyle=solid}
    InetSocketAddress httpBindAddress =
      NetUtils.createSocketAddr(httpBindAddressStr,
        YarnConfiguration.DEFAULT_NM_WEBAPP_PORT,
        YarnConfiguration.NM_WEBAPP_ADDRESS);
    try {
      //      this.hostName = InetAddress.getLocalHost().getCanonicalHostName();
      this.httpPort = httpBindAddress.getPort();
{code}
                
> When 0 is provided as port number in yarn.nodemanager.webapp.address, NMs 
> webserver component picks up random port, NM keeps on Reporting 0 port to RM
> ------------------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-3532
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-3532
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: nodemanager
>    Affects Versions: 0.23.1
>            Reporter: Karam Singh
>            Priority: Critical
>         Attachments: MAPREDUCE-3532.patch
>
>
> I tried following -:
> yarn.nodemanager.address=0.0.0.0:0
> yarn.nodemanager.webapp.address=0.0.0.0:0
> yarn.nodemanager.localizer.address=0.0.0.0:0
> mapreduce.shuffle.port=0
> When 0 is provided as number in yarn.nodemanager.webapp.address. 
> NM instantiate WebServer as 0 piort e.g.
> {code}
> 2011-12-08 11:33:02,467 INFO 
> org.apache.hadoop.yarn.server.nodemanager.webapp.WebServer: Instantiating 
> NMWebApp at 0.0.0.0:0
> {code}
> After that WebServer pick up some random port e.g.
> {code}
> 2011-12-08 11:33:02,562 INFO org.apache.hadoop.http.HttpServer: Jetty bound 
> to port 36272
> 2011-12-08 11:33:02,562 INFO org.mortbay.log: jetty-6.1.26
> 2011-12-08 11:33:02,831 INFO org.mortbay.log: Started 
> [email protected]:36272
> 2011-12-08 11:33:02,831 INFO org.apache.hadoop.yarn.webapp.WebApps: Web app 
> /node started at 36272
> {code}
> And NM WebServer responds correctly but
>  RM's cluster/Nodes page shows the following -:
> {code}
> /Rack RUNNING NM:57963 NM:0 Healthy 8-Dec-2011 11:33:01 Healthy 8 12 GB 0 KB
> {code}
> Whereas NM:0 is not clickable.
> Seems even NM's webserver pick random port but it never gets updated and so 
> NM report 0 as HTTP port to RM causing NM Hyperlinks un-clickable
> But verified that MR job runs successfully with random.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to