You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@whirr.apache.org by "Periya.Data" <pe...@gmail.com> on 2011/12/07 23:58:31 UTC
trouble reaching ec-2 instance
Hi All,
I am seeing a problem which has been reported earlier a few times, but,
I do not know the solution yet. I am unable to run "hadoop fs -ls /"
successfully "most" of the time from my laptop.
Some specs:
-----------------------
- My laptop :
Ubuntu 11.10 Oneiric Ocelot
Hadoop-0.20.2-cdh3u2
Apache Whirr 0.5.0-cdh3u2
EC2 instance: Ubuntu 10.04 32-bit LTS
*On term-1, I ran the proxy:*
sri@PeriyaData:~$ . ~/.whirr/HadoopDemo/hadoop-proxy.sh
Running proxy to Hadoop cluster at
ec2-174-129-76-176.compute-1.amazonaws.com. Use Ctrl-c to quit.
Warning: Permanently added
'ec2-174-129-76-176.compute-1.amazonaws.com,174.129.76.176'
(RSA) to the list of known hosts.
*On term-2, I exported conf-dir and did "hadoop fs - ls /"*
sri@PeriyaData:~$ export HADOOP_CONF_DIR=~/.whirr/JTVHadoopDemo/
sri@PeriyaData:~$ hadoop fs -ls /
11/12/07 13:43:18 WARN conf.Configuration: DEPRECATED: hadoop-site.xml
found in the classpath. Usage of hadoop-site.xml is deprecated. Instead use
core-site.xml, mapred-site.xml and hdfs-site.xml to override properties of
core-default.xml, mapred-default.xml and hdfs-default.xml respectively
Bad connection to FS. command aborted. exception: Call to
ec2-174-129-76-176.compute-1.amazonaws.com/174.129.76.176:8020 failed on
local exception: java.io.EOFException
sri@PeriyaData:~$
I tried the suggestion given at:
http://mail-archives.apache.org/mod_mbox/incubator-whirr-user/201106.mbox/%3C3A20A174-8D8B-4F84-80AF-C81A01B3732A@mortardata.com%3E.
Still did not work (after replacing host with external IP address).
According to this user who got it working, he mentions that the version of
hadoop instances must match...
Occasionally, for a different AMI, I get the "INFO ipc.Client: Retrying
connect to server:" issue. This has been reported here:
http://mail-archives.apache.org/mod_mbox/incubator-whirr-user/201107.mbox/%3CCAGtpQ96617csFRPL2ikq8p_W=TtXiod1SpQBe8WPNa4e5nq95g@mail.gmail.com%3E
Attached: Namenode and Jobtracker logs.
The reason why initially mentioned "mostly" is because I was able to run
this command successfully one time, but, since I was switching AMIs, I do
not remember.
Also, along the same lines, I am unable to use proxy on Firefox..even after
I ran the hadoop-proxy.sh and loaded the PAC file. Unable to view the
"browse the filesystem" link from the dfshealth.jsp page. Not sure if this
problem is a related one.
Thanks,
PD.
Re: trouble reaching ec-2 instance
Posted by Andrei Savu <sa...@gmail.com>.
Inline.
On Thu, Dec 8, 2011 at 12:58 AM, Periya.Data <pe...@gmail.com> wrote:
> Hi All,
> I am seeing a problem which has been reported earlier a few times,
> but, I do not know the solution yet. I am unable to run "hadoop fs -ls /"
> successfully "most" of the time from my laptop.
>
> Some specs:
> -----------------------
> - My laptop :
> Ubuntu 11.10 Oneiric Ocelot
> Hadoop-0.20.2-cdh3u2
> Apache Whirr 0.5.0-cdh3u2
>
The latest release is 0.6.0. You can download from here:
http://www.apache.org/dyn/closer.cgi/incubator/whirr/
>
> EC2 instance: Ubuntu 10.04 32-bit LTS
> *On term-1, I ran the proxy:*
> sri@PeriyaData:~$ . ~/.whirr/HadoopDemo/hadoop-proxy.sh
> Running proxy to Hadoop cluster at
> ec2-174-129-76-176.compute-1.amazonaws.com. Use Ctrl-c to quit.
> Warning: Permanently added 'ec2-174-129-76-176.compute-1.amazonaws.com,174.129.76.176'
> (RSA) to the list of known hosts.
>
>
> *On term-2, I exported conf-dir and did "hadoop fs - ls /"*
> sri@PeriyaData:~$ export HADOOP_CONF_DIR=~/.whirr/JTVHadoopDemo/
> sri@PeriyaData:~$ hadoop fs -ls /
> 11/12/07 13:43:18 WARN conf.Configuration: DEPRECATED: hadoop-site.xml
> found in the classpath. Usage of hadoop-site.xml is deprecated. Instead use
> core-site.xml, mapred-site.xml and hdfs-site.xml to override properties of
> core-default.xml, mapred-default.xml and hdfs-default.xml respectively
> Bad connection to FS. command aborted. exception: Call to
> ec2-174-129-76-176.compute-1.amazonaws.com/174.129.76.176:8020 failed on
> local exception: java.io.EOFException
Try with Whirr 0.6.0 - I am not sure if 0.5.0 install the latest CDH
release.
> Attached: Namenode and Jobtracker logs.
>
> The reason why initially mentioned "mostly" is because I was able to run
> this command successfully one time, but, since I was switching AMIs, I do
> not remember.
>
The AMI should be fine. I suspect this is either a version mistmatch or a
local environment issue.
>
Also, along the same lines, I am unable to use proxy on Firefox..even after
> I ran the hadoop-proxy.sh and loaded the PAC file. Unable to view the
> "browse the filesystem" link from the dfshealth.jsp page. Not sure if this
> problem is a related one.
>
How does a broken link look like?
>
> Thanks,
> PD.
>
>