hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Roman Shaposhnik <...@apache.org>
Subject Re: Running preliminary testing on HBase 0.92/Hadoop 0.22
Date Thu, 03 Nov 2011 00:27:33 GMT
Guys, thanks a million for the help!

It was a combination of a protocol version change and old client
(darn -- I was only behind by a week or so!)

Goes to show how fun it is to track trunk ;-)

Thanks,
Roman.

On Wed, Nov 2, 2011 at 3:11 PM, Stack <stack@duboce.net> wrote:
> On Wed, Nov 2, 2011 at 12:38 PM, Ted Yu <yuzhihong@gmail.com> wrote:
>>> On Wed, Nov 2, 2011 at 12:22 PM, Roman Shaposhnik <rvs@apache.org> wrote:
>>>
>>>> On Wed, Nov 2, 2011 at 10:42 AM, Stack <stack@duboce.net> wrote:
>>>> > In HBaseClient, its here:
>>>> >
>>>> >      if (remoteId.getAddress().isUnresolved()) {
>>>> >        throw new UnknownHostException("unknown host: " +
>>>> >
>>>> remoteId.getAddress().getHostName());
>>>> >      }
>>>> >
>>>> >
>>>> > getAddress is an InetSocketAddress..... and javadoc for isUnresolved:
>>>> >
>>>> >  "....true if the hostname couldn't be resolved into an InetAddress."
>>>> >
>>>> > Its kinda baffling Roman especially if other clients -- shell -- can
>>>> > get to master fine.
>>>>
>>>> Got that. Now, here's the odd part. When I connect to the VM with
>>>> a debugger and dump the value of
>>>>     remoteId.address.hostname
>>>> here's what I get:
>>>>     "\u0000\u0000ip-10-118-254-245.ec2.internal"
>>>>
>>>> That, of course, explains why such an address can't be resolved,
>>>> but do you guys have any idea what could put those \u0000\u0000
>>>> in front of the actual hostname?
>>>>
>
> hbase-4300 added versioning to the ServerName that is stored up in zk.
>  Version is a short on the front of the name.  The 'first' VERSION is
> 0.  Looking in MasterAddressTracker in 0.92, I see it doing this when
> you call masterAddressTracker#getMasterAddress (in HConnectionManager
> inside getMaster): ServerName.parseVersionedServerName(bytes);  This
> should be stripping those version bytes off the front of the name.
> You are up-to-date Roman?
>
> St.Ack
>

Mime
View raw message