hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Divij Durve <divij.t...@gmail.com>
Subject Re: Datanode Cannot Connect To The Server
Date Thu, 16 Jul 2009 03:05:01 GMT
No i had no problem at all it works perfectly.

On Wed, Jul 15, 2009 at 7:14 PM, Boyu Zhang <boyuzhang35@gmail.com> wrote:

> I just tried the version 0.20.0 and I followed the link.
> I can get pseudo distributed version started, but I still cannot get the
> cluster started.
>
> It is the same error. The datanode cannot connect to the namenode, and the
> task tracker cannot connect to the job tracker. I start to think that maybe
> the network ports are blocked somehow. Did you have the same problems when
> you set up the cluster?
>
> Thanks a lot!
>
> On Wed, Jul 15, 2009 at 4:41 PM, Divij Durve <divij.tech@gmail.com> wrote:
>
> > Ya i agree i was using a lower version and it was a lot of trouble to get
> > things to work. It took quite a lot of effort to get anything going. with
> > 0.20.0 i got it up and running in about 20 min
> >
> >
> > On Wed, Jul 15, 2009 at 4:24 PM, Boyu Zhang <boyuzhang35@gmail.com>
> wrote:
> >
> > > I am using the version 0.18.3 right now, the configurations files you
> > > mentioned are available in the version 0.20. Anyway, I will try the
> > version
> > > 0.20, this version now is just painful, thanks  a lot!
> > >
> > >
> > > On Wed, Jul 15, 2009 at 4:17 PM, Divij Durve <divij.tech@gmail.com>
> > wrote:
> > >
> > > > instead of using hadoop-site.xml try using the individual .xml files
> > > > *conf/core-site.xml*, *conf/hdfs-site.xml* and *conf/mapred-site.xml
> > > > It might help. I used hadoop-site.xml and had a lot of trouble with
> it.
> > I
> > > > moved on and used these and everything is working now.
> > > >
> > > >
> > > >
> > >
> >
> http://hadoop.apache.org/core/docs/current/cluster_setup.html#Site+Configuration
> > > >
> > > > thats a link to  the site that should be useful. its the cluster
> setup
> > > just
> > > > follow it carefully and u should be good to go.
> > > >
> > > > *
> > > > On Wed, Jul 15, 2009 at 4:09 PM, Boyu Zhang <boyuzhang35@gmail.com>
> > > wrote:
> > > >
> > > > > Thank you Divij,
> > > > >
> > > > > The port I used on the nameserver and on the slave is the same:
> 1111.
> > > > > Actually, the hadoop-site.xml is the same one on both machines. is
> > that
> > > > > correct? Thanks a lot!
> > > > >
> > > > > On Wed, Jul 15, 2009 at 4:04 PM, Divij Durve <divij.tech@gmail.com
> >
> > > > wrote:
> > > > >
> > > > > > u might want to have a look at the port that is being used on
the
> > > > server
> > > > > > and
> > > > > > the port that the slave is trying to communicate on.
> > > > > >
> > > > > > On Wed, Jul 15, 2009 at 3:57 PM, Boyu Zhang <
> boyuzhang35@gmail.com
> > >
> > > > > wrote:
> > > > > >
> > > > > > > Dear all,
> > > > > > >
> > > > > > > I am trying to set up a cluster of two machines using Hadoop.
> One
> > > > > machine
> > > > > > > is
> > > > > > > both namenode and jobtracker, the other machine is the
datanode
> > and
> > > > > > > tasktracker.
> > > > > > >
> > > > > > > I set up passwd-less ssh both directions.
> > > > > > >
> > > > > > >
> > > > > > > The comments I used are:
> > > > > > >  > bin/hadoop namenode -format
> > > > > > >  > bin/start-dfs.sh
> > > > > > >   starting namenode, logging to /data/zhang/hadoop/bin/../
> > > > > > > logs/hadoop-zhang-namenode-norway.out
> > > > > > >   datanode-machine name <http://sweden.gcl.cis.udel.edu/>:
> > > starting
> > > > > > > datanode, logging to
> > > > > > > /data/zhang/hadoop/bin/../logs/hadoop-zhang-datanode-sweden.out
> > > > > > >   datanode-machine name <http://sweden.gcl.cis.udel.edu/>:
> > > starting
> > > > > > > secondarynamenode, logging to
> > > > > > >
> > > > >
> > >
> /data/zhang/hadoop/bin/../logs/hadoop-zhang-secondarynamenode-sweden.out
> > > > > > >
> > > > > > >
> > > > > > > But when I check out the log files on the datanode machine,
I
> got
> > > the
> > > > > > > following error:
> > > > > > >
> > > > > > > 2009-07-15 15:45:54,226 INFO org.apache.hadoop.ipc.Client:
> > Retrying
> > > > > > connect
> > > > > > > to server:namenode-machine name:port. Already tried 2 time(s).
> > > > > > >
> > > > > > >
> > > > > > > And I the datanode cannot connect to the namenode, I cannot
> > operate
> > > > on
> > > > > > the
> > > > > > > hdfs.
> > > > > > >
> > > > > > > Could anyone can help me out?
> > > > > > >  Thanks a lot!
> > > > > > >
> > > > > > >
> > > > > > > Boyu
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>

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