error connecting to port 50002

[Mark Kerzner <ma...@gmail.com>]

Hi,

I often get this error

09/11/30 20:30:42 INFO ipc.Client: Retrying connect to server:
domU-12-31-39-00-DD-84.compute-1.internal/10.254.226.114:50002.

trying to run Hadoop jobs. This is  0.19 Apache distro on Amazon EC2.

Often I don't get it!

Either it's me or It looks like 0.19 is unstable.

Thank you,
Mark

Re: error connecting to port 50002

[Mark Kerzner <ma...@gmail.com>]

In the final analysis, the Apache Hadoop 0.19 cluster on EC2 sometimes has
problems running after first start-up, and almost always has problems after
running a job. Rebooting fixes it.

Do other people observe the same behavior? I have installed some additional
software on the Amazon Hadoop AMI, so it could be interfering also...

Thank you,
Mark

On Mon, Nov 30, 2009 at 8:09 PM, Mark Kerzner <ma...@gmail.com> wrote:

> After I reboot the cluster, 50002 begins to answer. It seems like the
> machines come up in the wrong way
>
> On Mon, Nov 30, 2009 at 7:32 PM, Mark Kerzner <ma...@gmail.com>wrote:
>
>> Hi,
>>
>> I often get this error
>>
>> 09/11/30 20:30:42 INFO ipc.Client: Retrying connect to server:
>> domU-12-31-39-00-DD-84.compute-1.internal/10.254.226.114:50002.
>>
>> trying to run Hadoop jobs. This is  0.19 Apache distro on Amazon EC2.
>>
>> Often I don't get it!
>>
>> Either it's me or It looks like 0.19 is unstable.
>>
>> Thank you,
>> Mark
>>
>
>

Re: error connecting to port 50002

[Mark Kerzner <ma...@gmail.com>]

After I reboot the cluster, 50002 begins to answer. It seems like the
machines come up in the wrong way

On Mon, Nov 30, 2009 at 7:32 PM, Mark Kerzner <ma...@gmail.com> wrote:

> Hi,
>
> I often get this error
>
> 09/11/30 20:30:42 INFO ipc.Client: Retrying connect to server:
> domU-12-31-39-00-DD-84.compute-1.internal/10.254.226.114:50002.
>
> trying to run Hadoop jobs. This is  0.19 Apache distro on Amazon EC2.
>
> Often I don't get it!
>
> Either it's me or It looks like 0.19 is unstable.
>
> Thank you,
> Mark
>