Hi,
   how can thrift server connect to hbase, through zookeeper?  how to
config it?
Thanks!
beatls

Hi Xiang,
please don't send question like this ... nobody can't response message like
this.
This is one of questions were response can be: "yes, you can, go it"
I write this message only because we have many client for access to HBase
via thrift and i want share our experience and take some new ideas.

What's your client language?

Here is many options, for exaple we have enabled
hbase.regionserver.export.thrift
and we use all region servers from client side (this is maybe bad, i don'
know)

LesTR
--
send from lower keyboard, please excuse my typos

**
2012/12/22 Xiang Hua <[EMAIL PROTECTED]>

> Hi,
>    how can thrift server connect to hbase, through zookeeper?  how to
> config it?
>
>
> Thanks!
>
>
> beatls
>

--
Save The World - http://www.worldcommunitygrid.org/
http://www.worldcommunitygrid.org/stat/viewMemberInfo.do?userName=LesTR

LesTR

Hi,
       we have a c++ client, and the thrift server was installed on the
Ndatanode ®ionservers.
      ( from the official document we know that it is not recommended for
the thrift colocated with nn &jobtracker).
      as we understand the processs from client to hbase shown below:
     client--->thrift server----->zookeeper---->hbase.
     my quesiton: can thrift server deploy on a stangalone server( do not
colocated with NN&Jobtracker);

     Many Thanks!
     beatls

On Sun, Dec 23, 2012 at 7:52 AM, Lukáš Drbal <[EMAIL PROTECTED]> wrote:

> Hi Xiang,
> please don't send question like this ... nobody can't response message like
> this.
> This is one of questions were response can be: "yes, you can, go it"
> I write this message only because we have many client for access to HBase
> via thrift and i want share our experience and take some new ideas.
>
> What's your client language?
>
> Here is many options, for exaple we have enabled
> hbase.regionserver.export.thrift
> and we use all region servers from client side (this is maybe bad, i don'
> know)
>
> LesTR
> --
> send from lower keyboard, please excuse my typos
>
>
>
> **
> 2012/12/22 Xiang Hua <[EMAIL PROTECTED]>
>
> > Hi,
> >    how can thrift server connect to hbase, through zookeeper?  how to
> > config it?
> >
> >
> > Thanks!
> >
> >
> > beatls
> >
>
>
>
> --
> Save The World - http://www.worldcommunitygrid.org/
> http://www.worldcommunitygrid.org/stat/viewMemberInfo.do?userName=LesTR
>
> LesTR
>

hi beatls，

for your question, the answer is YES!
if you just need a c++ client to access HBase,
just run thrift server like this:
in hbase home dir, run
bin/hbase-daemon.sh start thrift
or
bin/hbase-daemon.sh start thrift2

then you can using your c++ thrift client to connect.
I have some articles about  via thrift connect hbase in Chinese:
http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
http://abloz.com/2012/10/18/php-thrift-access-hbase-two.html
http://abloz.com/2012/05/31/example-thrift-installation.html

I'm using python and php connect hbase,not c++.
I also translate part of the HBase document to Chinese,url is:

http://abloz.com/hbase/book.html

for your reference.

Thanks.
Andy Zhou

2012/12/24 hua beatls <[EMAIL PROTECTED]>

> Hi,
>        we have a c++ client, and the thrift server was installed on the
> Ndatanode ®ionservers.
>       ( from the official document we know that it is not recommended for
> the thrift colocated with nn &jobtracker).
>       as we understand the processs from client to hbase shown below:
>      client--->thrift server----->zookeeper---->hbase.
>      my quesiton: can thrift server deploy on a stangalone server( do not
> colocated with NN&Jobtracker);
>
>      Many Thanks!
>
>
>      beatls
>
>
>
> On Sun, Dec 23, 2012 at 7:52 AM, Lukáš Drbal <[EMAIL PROTECTED]>
> wrote:
>
> > Hi Xiang,
> > please don't send question like this ... nobody can't response message
> like
> > this.
> > This is one of questions were response can be: "yes, you can, go it"
> > I write this message only because we have many client for access to HBase
> > via thrift and i want share our experience and take some new ideas.
> >
> > What's your client language?
> >
> > Here is many options, for exaple we have enabled
> > hbase.regionserver.export.thrift
> > and we use all region servers from client side (this is maybe bad, i don'
> > know)
> >
> > LesTR
> > --
> > send from lower keyboard, please excuse my typos
> >
> >
> >
> > **
> > 2012/12/22 Xiang Hua <[EMAIL PROTECTED]>
> >
> > > Hi,
> > >    how can thrift server connect to hbase, through zookeeper?  how to
> > > config it?
> > >
> > >
> > > Thanks!
> > >
> > >
> > > beatls
> > >
> >
> >
> >
> > --
> > Save The World - http://www.worldcommunitygrid.org/
> > http://www.worldcommunitygrid.org/stat/viewMemberInfo.do?userName=LesTR
> >
> > LesTR
> >
>

Hi Andy,
      you give me two ways to install (or configure) thriftserver , first
is showed @http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
     second way @
http://abloz.com/2012/05/31/example-thrift-installation.html, i prefer the
first one, but seems the second one if official?
      which one is better, or both works?

     Best R.

beatls

On Mon, Dec 24, 2012 at 7:49 PM, 周梦想 <[EMAIL PROTECTED]> wrote:

> hi beatls，
>
> for your question, the answer is YES!
> if you just need a c++ client to access HBase,
> just run thrift server like this:
> in hbase home dir, run
> bin/hbase-daemon.sh start thrift
> or
> bin/hbase-daemon.sh start thrift2
>
> then you can using your c++ thrift client to connect.
> I have some articles about  via thrift connect hbase in Chinese:
> http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
> http://abloz.com/2012/10/18/php-thrift-access-hbase-two.html
> http://abloz.com/2012/05/31/example-thrift-installation.html
>
> I'm using python and php connect hbase,not c++.
> I also translate part of the HBase document to Chinese,url is:
>
> http://abloz.com/hbase/book.html
>
> for your reference.
>
> Thanks.
> Andy Zhou
>
> 2012/12/24 hua beatls <[EMAIL PROTECTED]>
>
> > Hi,
> >        we have a c++ client, and the thrift server was installed on the
> > Ndatanode ®ionservers.
> >       ( from the official document we know that it is not recommended for
> > the thrift colocated with nn &jobtracker).
> >       as we understand the processs from client to hbase shown below:
> >      client--->thrift server----->zookeeper---->hbase.
> >      my quesiton: can thrift server deploy on a stangalone server( do not
> > colocated with NN&Jobtracker);
> >
> >      Many Thanks!
> >
> >
> >      beatls
> >
> >
> >
> > On Sun, Dec 23, 2012 at 7:52 AM, Lukáš Drbal <[EMAIL PROTECTED]>
> > wrote:
> >
> > > Hi Xiang,
> > > please don't send question like this ... nobody can't response message
> > like
> > > this.
> > > This is one of questions were response can be: "yes, you can, go it"
> > > I write this message only because we have many client for access to
> HBase
> > > via thrift and i want share our experience and take some new ideas.
> > >
> > > What's your client language?
> > >
> > > Here is many options, for exaple we have enabled
> > > hbase.regionserver.export.thrift
> > > and we use all region servers from client side (this is maybe bad, i
> don'
> > > know)
> > >
> > > LesTR
> > > --
> > > send from lower keyboard, please excuse my typos
> > >
> > >
> > >
> > > **
> > > 2012/12/22 Xiang Hua <[EMAIL PROTECTED]>
> > >
> > > > Hi,
> > > >    how can thrift server connect to hbase, through zookeeper?  how to
> > > > config it?
> > > >
> > > >
> > > > Thanks!
> > > >
> > > >
> > > > beatls
> > > >
> > >
> > >
> > >
> > > --
> > > Save The World - http://www.worldcommunitygrid.org/
> > >
> http://www.worldcommunitygrid.org/stat/viewMemberInfo.do?userName=LesTR
> > >
> > > LesTR
> > >
> >
>

Hi Andy,
    shoud i first install as
http://abloz.com/2012/05/31/example-thrift-installation.html, then
configure thrift as @
http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html. that's
first install thrift using tarball from apache.thirft.org, then configrue
and start it with './hbase-daemon.sh start thrift'.
    Best R.

    beatls

On Tue, Dec 25, 2012 at 9:29 PM, hua beatls <[EMAIL PROTECTED]> wrote:

> Hi Andy,
>       you give me two ways to install (or configure) thriftserver , first
> is showed @
> http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
>      second way @
> http://abloz.com/2012/05/31/example-thrift-installation.html, i prefer
> the first one, but seems the second one if official?
>       which one is better, or both works?
>
>      Best R.
>
> beatls
>
>
> On Mon, Dec 24, 2012 at 7:49 PM, 周梦想 <[EMAIL PROTECTED]> wrote:
>
>> hi beatls，
>>
>> for your question, the answer is YES!
>> if you just need a c++ client to access HBase,
>> just run thrift server like this:
>> in hbase home dir, run
>> bin/hbase-daemon.sh start thrift
>> or
>> bin/hbase-daemon.sh start thrift2
>>
>> then you can using your c++ thrift client to connect.
>> I have some articles about  via thrift connect hbase in Chinese:
>> http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
>> http://abloz.com/2012/10/18/php-thrift-access-hbase-two.html
>> http://abloz.com/2012/05/31/example-thrift-installation.html
>>
>> I'm using python and php connect hbase,not c++.
>> I also translate part of the HBase document to Chinese,url is:
>>
>> http://abloz.com/hbase/book.html
>>
>> for your reference.
>>
>> Thanks.
>> Andy Zhou
>>
>> 2012/12/24 hua beatls <[EMAIL PROTECTED]>
>>
>> > Hi,
>> >        we have a c++ client, and the thrift server was installed on the
>> > Ndatanode ®ionservers.
>> >       ( from the official document we know that it is not recommended
>> for
>> > the thrift colocated with nn &jobtracker).
>> >       as we understand the processs from client to hbase shown below:
>> >      client--->thrift server----->zookeeper---->hbase.
>> >      my quesiton: can thrift server deploy on a stangalone server( do
>> not
>> > colocated with NN&Jobtracker);
>> >
>> >      Many Thanks!
>> >
>> >
>> >      beatls
>> >
>> >
>> >
>> > On Sun, Dec 23, 2012 at 7:52 AM, Lukáš Drbal <[EMAIL PROTECTED]>
>> > wrote:
>> >
>> > > Hi Xiang,
>> > > please don't send question like this ... nobody can't response message
>> > like
>> > > this.
>> > > This is one of questions were response can be: "yes, you can, go it"
>> > > I write this message only because we have many client for access to
>> HBase
>> > > via thrift and i want share our experience and take some new ideas.
>> > >
>> > > What's your client language?
>> > >
>> > > Here is many options, for exaple we have enabled
>> > > hbase.regionserver.export.thrift
>> > > and we use all region servers from client side (this is maybe bad, i
>> don'
>> > > know)
>> > >
>> > > LesTR
>> > > --
>> > > send from lower keyboard, please excuse my typos
>> > >
>> > >
>> > >
>> > > **
>> > > 2012/12/22 Xiang Hua <[EMAIL PROTECTED]>
>> > >
>> > > > Hi,
>> > > >    how can thrift server connect to hbase, through zookeeper?  how
>> to
>> > > > config it?
>> > > >
>> > > >
>> > > > Thanks!
>> > > >
>> > > >
>> > > > beatls
>> > > >
>> > >
>> > >
>> > >
>> > > --
>> > > Save The World - http://www.worldcommunitygrid.org/
>> > >
>> http://www.worldcommunitygrid.org/stat/viewMemberInfo.do?userName=LesTR
>> > >
>> > > LesTR
>> > >
>> >
>>
>
>

Yes, the first article is the base to install and test thrift, it has
nothing to do with HBase.
And the second one is how to use thrift  on HBase.

2012/12/25 hua beatls <[EMAIL PROTECTED]>

> Hi Andy,
>     shoud i first install as
> http://abloz.com/2012/05/31/example-thrift-installation.html, then
> configure thrift as @
> http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html. that's
> first install thrift using tarball from apache.thirft.org, then configrue
> and start it with './hbase-daemon.sh start thrift'.
>
>
>     Best R.
>
>     beatls
>
> On Tue, Dec 25, 2012 at 9:29 PM, hua beatls <[EMAIL PROTECTED]> wrote:
>
> > Hi Andy,
> >       you give me two ways to install (or configure) thriftserver , first
> > is showed @
> > http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
> >      second way @
> > http://abloz.com/2012/05/31/example-thrift-installation.html, i prefer
> > the first one, but seems the second one if official?
> >       which one is better, or both works?
> >
> >      Best R.
> >
> > beatls
> >
> >
> > On Mon, Dec 24, 2012 at 7:49 PM, 周梦想 <[EMAIL PROTECTED]> wrote:
> >
> >> hi beatls，
> >>
> >> for your question, the answer is YES!
> >> if you just need a c++ client to access HBase,
> >> just run thrift server like this:
> >> in hbase home dir, run
> >> bin/hbase-daemon.sh start thrift
> >> or
> >> bin/hbase-daemon.sh start thrift2
> >>
> >> then you can using your c++ thrift client to connect.
> >> I have some articles about  via thrift connect hbase in Chinese:
> >> http://abloz.com/2012/06/01/python-operating-hbase-thrift-to.html
> >> http://abloz.com/2012/10/18/php-thrift-access-hbase-two.html
> >> http://abloz.com/2012/05/31/example-thrift-installation.html
> >>
> >> I'm using python and php connect hbase,not c++.
> >> I also translate part of the HBase document to Chinese,url is:
> >>
> >> http://abloz.com/hbase/book.html
> >>
> >> for your reference.
> >>
> >> Thanks.
> >> Andy Zhou
> >>
> >> 2012/12/24 hua beatls <[EMAIL PROTECTED]>
> >>
> >> > Hi,
> >> >        we have a c++ client, and the thrift server was installed on
> the
> >> > Ndatanode ®ionservers.
> >> >       ( from the official document we know that it is not recommended
> >> for
> >> > the thrift colocated with nn &jobtracker).
> >> >       as we understand the processs from client to hbase shown below:
> >> >      client--->thrift server----->zookeeper---->hbase.
> >> >      my quesiton: can thrift server deploy on a stangalone server( do
> >> not
> >> > colocated with NN&Jobtracker);
> >> >
> >> >      Many Thanks!
> >> >
> >> >
> >> >      beatls
> >> >
> >> >
> >> >
> >> > On Sun, Dec 23, 2012 at 7:52 AM, Lukáš Drbal <[EMAIL PROTECTED]>
> >> > wrote:
> >> >
> >> > > Hi Xiang,
> >> > > please don't send question like this ... nobody can't response
> message
> >> > like
> >> > > this.
> >> > > This is one of questions were response can be: "yes, you can, go it"
> >> > > I write this message only because we have many client for access to
> >> HBase
> >> > > via thrift and i want share our experience and take some new ideas.
> >> > >
> >> > > What's your client language?
> >> > >
> >> > > Here is many options, for exaple we have enabled
> >> > > hbase.regionserver.export.thrift
> >> > > and we use all region servers from client side (this is maybe bad, i
> >> don'
> >> > > know)
> >> > >
> >> > > LesTR
> >> > > --
> >> > > send from lower keyboard, please excuse my typos
> >> > >
> >> > >
> >> > >
> >> > > **
> >> > > 2012/12/22 Xiang Hua <[EMAIL PROTECTED]>
> >> > >
> >> > > > Hi,
> >> > > >    how can thrift server connect to hbase, through zookeeper?  how
> >> to
> >> > > > config it?
> >> > > >
> >> > > >
> >> > > > Thanks!
> >> > > >
> >> > > >
> >> > > > beatls
> >> > > >
> >> > >
> >> > >
> >> > >
> >> > > --
> >> > > Save The World - http://www.worldcommunitygrid.org/
> >> > >
> >> http://www.worldcommunitygrid.org/stat/viewMemberInfo.do?userName=LesTR
> >> > >
> >> > > LesTR
> >> > >
> >> >
> >>

Hi Andy，
     i have a question about thrift,does thrift server connect to hbase
through zookeeper?
     i read the thrift log and find thrift is assign 'request' to different
regionserver.
     below is the log:
     2012-12-27 15:39:27,924 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
Cached location for BT_NET_LOG_000,0000189094114602012122620
200107000031857,1356593966233.d20458ecf526a932f602af63002b290e. is
hadoop1:60020
2012-12-27 15:39:27,924 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
Cached location for BT_NET_LOG_000,0000189098696122012122921
000107000031857,1356593966233.c1f127dc9de117605ea332f27b0b3775. is
hadoop1:60020
2012-12-27 15:39:28,563 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
Scanning .META. starting at
row=BT_NET_LOG_000,0000189094114602012122620200107000031857,00000
000000000 for max=10 rows using
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
2012-12-27 15:45:14,850 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
Removed BT_NET_LOG_000,0000153910000002012120115134001000000
272,1356589842295.5c84298f6889734514903fffc9582689. for
tableName=BT_NET_LOG_000 from cache because of
0000189002087912012120100000107000031857
2012-12-27 15:45:14,853 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
Retry 1, sleep for 1000ms!
2012-12-27 15:45:15,379 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
Scanning .META. starting at
row=BT_NET_LOG_000,0000153910000002012120115134001000000272,00000
000000000 for max=10 rows using
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
2012-12-27 15:45:15,384 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
locateRegionInMeta parentTable=.META., metaLocation={region.META.,,1.1028785192, hostname=hadoop1, port=60020}, attempt=0 of 10
failed; retrying after sleep of 1000 because: the only available region for
the required row is a
split parent, the daughters should be online soon:
BT_NET_LOG_000,0000153910000002012120115134001000000272,1356589842295.5c84298f6889734514903fffc9582689.
2012-12-27 15:45:15,856 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
Scanning .META. starting at
row=BT_NET_LOG_000,0000189001962242012122322200106000000559,00000
000000000 for max=10 rows using
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
2012-12-27 15:45:15,859 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
Cached location for BT_NET_LOG_000,0000189001962242012122322
200106000000559,1356594314689.23d2fce6f5b6912e39eb7bdf22a069b3. is
hadoop3:60020
2012-12-27 15:45:16,387 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
Scanning .META. starting at
row=BT_NET_LOG_000,0000189001962242012122322200106000000559,00000
000000000 for max=10 rows using
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
2012-12-27 15:45:17,638 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
Scanning .META. starting at
row=BT_NET_LOG_000,0000153910000002012120115134001000000272,00000
000000000 for max=10 rows using
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
2012-12-27 15:45:17,642 DEBUG
org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
Cached location for BT_NET_LOG_000,0000153910000002012120115
134001000000272,1356594314689.91a440fc38a1d4d9b0afbf0ebf33e7a8. is
hadoop3:60020

    seems thrift assign request to hadoop1 and hadoop3, but i still have a
hadoop2 which seems not included in this circulation.
    any suggestion?
    Thanks!

   beatls

On Wed, Dec 26, 2012 at 2:52 PM, 周梦想 <[EMAIL PROTECTED]> wrote:

> Yes, the first article is the base to install and test thrift, it has
> nothing to do with HBase.
> And the second one is how to use thrift  on HBase.
>
> 2012/12/25 hua beatls <[EMAIL PROTECTED]>

Hi hua,
The zookeeper is used by HBase for tow main purpose, one is manging every
region server state, the other is managing --ROOT-- table updated by
HMaster.  So most HBase operation will keep touch with zookeeper,  the
thrift server is not an exception.

2012/12/27 hua beatls <[EMAIL PROTECTED]>

> Hi Andy，
>      i have a question about thrift,does thrift server connect to hbase
> through zookeeper?
>      i read the thrift log and find thrift is assign 'request' to different
> regionserver.
>      below is the log:
>      2012-12-27 15:39:27,924 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> Cached location for BT_NET_LOG_000,0000189094114602012122620
> 200107000031857,1356593966233.d20458ecf526a932f602af63002b290e. is
> hadoop1:60020
> 2012-12-27 15:39:27,924 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> Cached location for BT_NET_LOG_000,0000189098696122012122921
> 000107000031857,1356593966233.c1f127dc9de117605ea332f27b0b3775. is
> hadoop1:60020
> 2012-12-27 15:39:28,563 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
> Scanning .META. starting at
> row=BT_NET_LOG_000,0000189094114602012122620200107000031857,00000
> 000000000 for max=10 rows using
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
> 2012-12-27 15:45:14,850 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> Removed BT_NET_LOG_000,0000153910000002012120115134001000000
> 272,1356589842295.5c84298f6889734514903fffc9582689. for
> tableName=BT_NET_LOG_000 from cache because of
> 0000189002087912012120100000107000031857
> 2012-12-27 15:45:14,853 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> Retry 1, sleep for 1000ms!
> 2012-12-27 15:45:15,379 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
> Scanning .META. starting at
> row=BT_NET_LOG_000,0000153910000002012120115134001000000272,00000
> 000000000 for max=10 rows using
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
> 2012-12-27 15:45:15,384 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> locateRegionInMeta parentTable=.META., metaLocation={region> .META.,,1.1028785192, hostname=hadoop1, port=60020}, attempt=0 of 10
> failed; retrying after sleep of 1000 because: the only available region for
> the required row is a
> split parent, the daughters should be online soon:
>
> BT_NET_LOG_000,0000153910000002012120115134001000000272,1356589842295.5c84298f6889734514903fffc9582689.
> 2012-12-27 15:45:15,856 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
> Scanning .META. starting at
> row=BT_NET_LOG_000,0000189001962242012122322200106000000559,00000
> 000000000 for max=10 rows using
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
> 2012-12-27 15:45:15,859 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> Cached location for BT_NET_LOG_000,0000189001962242012122322
> 200106000000559,1356594314689.23d2fce6f5b6912e39eb7bdf22a069b3. is
> hadoop3:60020
> 2012-12-27 15:45:16,387 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
> Scanning .META. starting at
> row=BT_NET_LOG_000,0000189001962242012122322200106000000559,00000
> 000000000 for max=10 rows using
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
> 2012-12-27 15:45:17,638 DEBUG org.apache.hadoop.hbase.client.MetaScanner:
> Scanning .META. starting at
> row=BT_NET_LOG_000,0000153910000002012120115134001000000272,00000
> 000000000 for max=10 rows using
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation@52313a4e
> 2012-12-27 15:45:17,642 DEBUG
>
> org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation:
> Cached location for BT_NET_LOG_000,0000153910000002012120115
> 134001000000272,1356594314689.91a440fc38a1d4d9b0afbf0ebf33e7a8. is
maybe your region data just asigned to hadoop1 and hadoop3?  you could
check hadoop2's state, and increase data and load to test.
Good luck!

Andy