hadoop-hdfs-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aaron Eng <a...@maprtech.com>
Subject Re: Hadoop on EC2 Managing Internal/External IPs
Date Fri, 24 Aug 2012 03:52:51 GMT
Hi Igor,

I don't think theres anything in Hadoop thats going to allow you to have an
internal IP assigned to a machines network interface and to have it
advertise the external IP.  Even if that were in place, you'd then have to
differentiate between requests coming from the other nodes in the cluster
vs. from external networks.  If your nodes were advertising their external
IP to other nodes, then its likely your network traffic would have to
traverse many more hops to be NAT'd and then come back.  The whole thing is
messy and would require a bunch of network configs that are probably going
to be a hassle to manage.

Also, I don't think you want to have your nodes exposed to the internet
since Hadoop traffic is unencrypted unless you've done something special to
encrypt it.  Even if you restrict your nodes to respond only to a specific
IP, you can't guarantee that someone else can't take over your IP from a
security perspective.  The VPN gateway offers both security and ease of

On Thu, Aug 23, 2012 at 8:09 PM, igor Finkelshteyn <iefinkel@gmail.com>wrote:

> That would work, but wouldn't a much simpler solution just be to force the
> machines in the cluster to always pass around their external FQDNs, since
> those will properly resolve to the internal or external IP depending on
> what machine is asking? Is there no way to just do that?
> On Aug 23, 2012, at 8:02 PM, Aaron Eng wrote:
> Hi Igor,
> Amazon offers a service where you can have a VPN gateway on your network
> that leads directly back to the network where youre instances are at.  So
> that 10.123.x.x subnet would be connected off of the VPN gateway on your
> network and you'd set up your routers/routing to push traffic for that
> subnet at the gateway.
> On Thu, Aug 23, 2012 at 12:34 PM, igor Finkelshteyn <iefinkel@gmail.com>wrote:
>> Hi,
>> I'm currently setting up a Hadoop cluster on EC2, and everything works
>> just fine when accessing the cluster from inside EC2, but as soon as I try
>> to do something like upload a file from an external client, I get timeout
>> errors like:
>> 12/08/23 12:06:16 ERROR hdfs.DFSClient: Failed to close file
>> /user/some_file._COPYING_
>> java.net.SocketTimeoutException: 65000 millis timeout while waiting for
>> channel to be ready for connect. ch :
>> java.nio.channels.SocketChannel[connection-pending remote=/10.123.x.x:50010]
>> What's clearly happening is my NameNode is resolving my DataNode's IPs to
>> their internal EC2 values instead of their external values, and then
>> sending along the internal IP to my external client, which is obviously
>> unable to reach those. I'm thinking this must be a common problem. How do
>> other people deal with it? Is there a way to just force my name node to
>> send along my DataNode's hostname instead of IP, so that the hostname can
>> be resolved properly from whatever box will be sending files?
>> Eli

View raw message