When I try to access HBase from a cluster node which is not the
zookeeper server I have the following problem:

-bash-3.2$ hostname
node33
-bash-3.2$ hbase shell
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012

hbase(main):001:0> status

ERROR: org.apache.hadoop.hbase.ZooKeeperConnectionException: HBase is
able to connect to ZooKeeper but the connection closes immediately.
This could be a sign that the server has too many connections (30 is
the default). Consider inspecting your ZK server logs for that error
and then make sure you are reusing HBaseConfiguration as often as you
can. See HTable's javadoc for more information.

Here is some help for this command:
Show cluster status. Can be 'summary', 'simple', or 'detailed'. The
default is 'summary'. Examples:

  hbase> status
  hbase> status 'simple'
  hbase> status 'summary'
  hbase> status 'detailed'
hbase(main):002:0>
Now, if I log into the zookeeper server I can do:

-bash-3.2$ hostname
master
-bash-3.2$ hbase shell
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012

hbase(main):001:0> status
34 servers, 0 dead, 59.0882 average load

hbase(main):002:0>
What's going on? I suspect that HBase should be able to read/write
from any node that I can modify hdfs from.

Hi Ryan,

  Make sure you have the correct client configurations on the node you are
trying to access from.  You will need the hbase-site and the zoo.cfg to
make this work.

On Tue, Mar 12, 2013 at 4:47 PM, Ryan Compton <[EMAIL PROTECTED]>wrote:

> When I try to access HBase from a cluster node which is not the
> zookeeper server I have the following problem:
>
> -bash-3.2$ hostname
> node33
> -bash-3.2$ hbase shell
> HBase Shell; enter 'help<RETURN>' for list of supported commands.
> Type "exit<RETURN>" to leave the HBase Shell
> Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012
>
> hbase(main):001:0> status
>
> ERROR: org.apache.hadoop.hbase.ZooKeeperConnectionException: HBase is
> able to connect to ZooKeeper but the connection closes immediately.
> This could be a sign that the server has too many connections (30 is
> the default). Consider inspecting your ZK server logs for that error
> and then make sure you are reusing HBaseConfiguration as often as you
> can. See HTable's javadoc for more information.
>
> Here is some help for this command:
> Show cluster status. Can be 'summary', 'simple', or 'detailed'. The
> default is 'summary'. Examples:
>
>   hbase> status
>   hbase> status 'simple'
>   hbase> status 'summary'
>   hbase> status 'detailed'
>
>
> hbase(main):002:0>
>
>
> Now, if I log into the zookeeper server I can do:
>
> -bash-3.2$ hostname
> master
> -bash-3.2$ hbase shell
> HBase Shell; enter 'help<RETURN>' for list of supported commands.
> Type "exit<RETURN>" to leave the HBase Shell
> Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012
>
> hbase(main):001:0> status
> 34 servers, 0 dead, 59.0882 average load
>
> hbase(main):002:0>
>
>
> What's going on? I suspect that HBase should be able to read/write
> from any node that I can modify hdfs from.
>

--
Kevin O'Dell
Customer Operations Engineer, Cloudera

Have you seen Kevin's response ?

Cheers

On Tue, Mar 12, 2013 at 1:42 PM, Ryan Compton <[EMAIL PROTECTED]>wrote:

> When I try to access HBase from a cluster node which is not the
> zookeeper server I have the following problem:
>
> -bash-3.2$ hostname
> node33
> -bash-3.2$ hbase shell
> HBase Shell; enter 'help<RETURN>' for list of supported commands.
> Type "exit<RETURN>" to leave the HBase Shell
> Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012
>
> hbase(main):001:0> status
>
> ERROR: org.apache.hadoop.hbase.ZooKeeperConnectionException: HBase is
> able to connect to ZooKeeper but the connection closes immediately.
> This could be a sign that the server has too many connections (30 is
> the default). Consider inspecting your ZK server logs for that error
> and then make sure you are reusing HBaseConfiguration as often as you
> can. See HTable's javadoc for more information.
>
> Here is some help for this command:
> Show cluster status. Can be 'summary', 'simple', or 'detailed'. The
> default is 'summary'. Examples:
>
>   hbase> status
>   hbase> status 'simple'
>   hbase> status 'summary'
>   hbase> status 'detailed'
>
>
> hbase(main):002:0>
>
>
> Now, if I log into the zookeeper server I can do:
>
> -bash-3.2$ hostname
> master
> -bash-3.2$ hbase shell
> HBase Shell; enter 'help<RETURN>' for list of supported commands.
> Type "exit<RETURN>" to leave the HBase Shell
> Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012
>
> hbase(main):001:0> status
> 34 servers, 0 dead, 59.0882 average load
>
> hbase(main):002:0>
>
>
> What's going on? I suspect that HBase should be able to read/write
> from any node that I can modify hdfs from.
>

Yes, I am having success with it.

We use Cloudera and there's a button, "Generate Client Configuration",
which generated a .zip file with a very informative README.txt. The
trick is to extract that .zip and export HBASE_CONF_DIR :

hbase(main):002:0> exit
-bash-3.2$ hostname
node33
-bash-3.2$ export HBASE_CONF_DIR=/data/cm-gui/hbase-global-clientconfig
-bash-3.2$ /usr/lib/hbase/bin/hbase shell
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012

hbase(main):001:0> status
34 servers, 0 dead, 59.0882 average load
So far, this is working and I can get to the shell from other nodes. I
think the installation process for mapreduce/hdfs is the same (but the
guy who set it all up quit a few weeks ago).
On Tue, Mar 12, 2013 at 2:08 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> Have you seen Kevin's response ?
>
> Cheers
>
> On Tue, Mar 12, 2013 at 1:42 PM, Ryan Compton <[EMAIL PROTECTED]>wrote:
>
>> When I try to access HBase from a cluster node which is not the
>> zookeeper server I have the following problem:
>>
>> -bash-3.2$ hostname
>> node33
>> -bash-3.2$ hbase shell
>> HBase Shell; enter 'help<RETURN>' for list of supported commands.
>> Type "exit<RETURN>" to leave the HBase Shell
>> Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012
>>
>> hbase(main):001:0> status
>>
>> ERROR: org.apache.hadoop.hbase.ZooKeeperConnectionException: HBase is
>> able to connect to ZooKeeper but the connection closes immediately.
>> This could be a sign that the server has too many connections (30 is
>> the default). Consider inspecting your ZK server logs for that error
>> and then make sure you are reusing HBaseConfiguration as often as you
>> can. See HTable's javadoc for more information.
>>
>> Here is some help for this command:
>> Show cluster status. Can be 'summary', 'simple', or 'detailed'. The
>> default is 'summary'. Examples:
>>
>>   hbase> status
>>   hbase> status 'simple'
>>   hbase> status 'summary'
>>   hbase> status 'detailed'
>>
>>
>> hbase(main):002:0>
>>
>>
>> Now, if I log into the zookeeper server I can do:
>>
>> -bash-3.2$ hostname
>> master
>> -bash-3.2$ hbase shell
>> HBase Shell; enter 'help<RETURN>' for list of supported commands.
>> Type "exit<RETURN>" to leave the HBase Shell
>> Version 0.90.4-cdh3u3, r, Thu Jan 26 10:13:36 PST 2012
>>
>> hbase(main):001:0> status
>> 34 servers, 0 dead, 59.0882 average load
>>
>> hbase(main):002:0>
>>
>>
>> What's going on? I suspect that HBase should be able to read/write
>> from any node that I can modify hdfs from.
>>