headers
bharath vissapragada | 1 Aug 2009 06:11
Picon

Re: HBase schema design

See the example on the bottom of the page ..

http://hadoop.apache.org/hbase/docs/r0.19.3/api/overview-summary.html

It is very well documented ... and u can easily understand!

On Fri, Jul 31, 2009 at 2:26 PM, Ninad Raut <hbase.user.ninad@...>wrote:

> Use a Scanner instead of a get....
>
> 2009/7/31 <y_823910@...>
>
> > As to your suggestion, how do I complete following codes?
> > or any better codes would be great!
> >
> > HTable table = new HTable(config, "user");
> > Get g = new Get(Bytes.toBytes("U1"));
> > Result r = table.get(g);
> >
> > for loop---     //I don't know how many articles the user have.
> >   byte [] t = r.getValue(Bytes.toBytes("chapter"), Bytes.toBytes("?"));
> > //t1
> >   byte [] c = r.getValue(Bytes.toBytes("chapter"), Bytes.toBytes("?"));
> > //c1
> >   byte [] d = r.getValue(Bytes.toBytes("chapter"), Bytes.toBytes("?"));
> > //d1
> > end loop---
> >
> > Thank you very much
> >
(Continue reading)

Permalink | Reply | 1 comment |
headers
lei wang | 1 Aug 2009 06:50
Picon

The problem with hbase0.20 “Getting Started” tutorial

The hadoop 0.20.X move hadoop-dite.xml to core-site.xml and hdfs-site.xml,
etc.
The tutorial claims in requirement ：This version of HBase will only
run on Hadoop
0.20.x <http://hadoop.apache.org/core/releases.html>.
But in the “Fully-Distributed Operation” chapter，the tutorial still
mentions：

   - Add a pointer to your HADOOP_CONF_DIR to CLASSPATH in hbase-env.sh
   - Add a copy of hadoop-site.xml to ${HBASE_HOME}/conf, or
   - If only a small set of HDFS client configurations, add them to
   hbase-site.xml

 where is hadoop-site.xml ?
Permalink | Reply | 0 comments |
headers
lei wang | 1 Aug 2009 12:47
Picon

Stack: Must hbase0.20 go with zookeeper?

Stack's mail remind me us the hbase 0.20 candidate version,  when does hbase
0.20.0 release?
In the hbase 0.20.0 api overview, it mention zookeeper cluster
configuration, but i am just use hadoop in hbase 0.19.3 before, how can i do
with hbase 0.20.0? I don't want to use zookeeper as a plugin.
Permalink | Reply | 4 comments |
headers
lei wang | 1 Aug 2009 15:13
Picon

Re: hbase 0.20.0 Release Candidate 1 available for download

The hadoop 0.20.X move hadoop-dite.xml to core-site.xml and hdfs-site.xml,
etc.
The tutorial claims in requirement ：This version of HBase will only
run on Hadoop
0.20.x <http://hadoop.apache.org/core/releases.html>.
But in the “Fully-Distributed Operation” chapter，the tutorial still
mentions：

   - Add a pointer to your HADOOP_CONF_DIR to CLASSPATH in hbase-env.sh
   - Add a copy of hadoop-site.xml to ${HBASE_HOME}/conf, or
   - If only a small set of HDFS client configurations, add them to
   hbase-site.xml

 where is hadoop-site.xml ?

On Thu, Jul 30, 2009 at 5:04 PM, Jonathan Gray <jlist@...> wrote:

> It is now part of contrib.  Look in src/contrib/transactional
>
>
> Yair Even-Zohar wrote:
>
>> 1) The link is broken. Only the first part is needed:
>> http://people.apache.org/~stack/hbase-0.20.0-candidate-1/<http://people.apache.org/%7Estack/hbase-0.20.0-candidate-1/>
>> 2) I was just looking for the tableIndexed. It seems like the package is
>> missing. Shouldn't the source be under
>> src\java\org\apache\hadoop\hbase\client  ?  (at least according to the
>> api docs )
>>
>> Thanks
(Continue reading)

Permalink | Reply | 11 comments |
headers
Jean-Daniel Cryans | 1 Aug 2009 17:13
Picon
Favicon

Re: hbase 0.20.0 Release Candidate 1 available for download

This is indeed an omission, it has to be fixed before official release.

Also please don't dual post, your message "The problem with hbase0.20
“Getting Started” tutorial" is a copy paste.

J-D

On Sat, Aug 1, 2009 at 9:13 AM, lei wang<nutchmaillist@...> wrote:
> The hadoop 0.20.X move hadoop-dite.xml to core-site.xml and hdfs-site.xml,
> etc.
> The tutorial claims in requirement ：This version of HBase will only
> run on Hadoop
> 0.20.x <http://hadoop.apache.org/core/releases.html>.
> But in the “Fully-Distributed Operation” chapter，the tutorial still
> mentions：
>
>   - Add a pointer to your HADOOP_CONF_DIR to CLASSPATH in hbase-env.sh
>   - Add a copy of hadoop-site.xml to ${HBASE_HOME}/conf, or
>   - If only a small set of HDFS client configurations, add them to
>   hbase-site.xml
>
>  where is hadoop-site.xml ?
>
> On Thu, Jul 30, 2009 at 5:04 PM, Jonathan Gray <jlist@...> wrote:
>
>> It is now part of contrib.  Look in src/contrib/transactional
>>
>>
>> Yair Even-Zohar wrote:
>>
(Continue reading)

Permalink | Reply | 10 comments |
headers
Jean-Daniel Cryans | 1 Aug 2009 17:17
Picon
Favicon

Re: Stack: Must hbase0.20 go with zookeeper?

HBase official release will happen when a release candidate gets only
+1s. I think the current one may get sunk, we'll see.

Starting with hbase-0.20, as the doc says, it will start a Zookeeper
quorum for you unless you override that behavior, in which case you
must start it yourself. So ZK is mandatory and it is already contained
in the distribution.

J-D

On Sat, Aug 1, 2009 at 6:47 AM, lei wang<hbasemaillist@...> wrote:
> Stack's mail remind me us the hbase 0.20 candidate version,  when does hbase
> 0.20.0 release?
> In the hbase 0.20.0 api overview, it mention zookeeper cluster
> configuration, but i am just use hadoop in hbase 0.19.3 before, how can i do
> with hbase 0.20.0? I don't want to use zookeeper as a plugin.
>
Permalink | Reply | 3 comments |
headers
lei wang | 1 Aug 2009 17:57
Picon

Re: hbase 0.20.0 Release Candidate 1 available for download

ok，thanks.
I am so sorry to dual the post.

On Sat, Aug 1, 2009 at 11:13 PM, Jean-Daniel Cryans <jdcryans@...>wrote:

> This is indeed an omission, it has to be fixed before official release.
>
> Also please don't dual post, your message "The problem with hbase0.20
> “Getting Started” tutorial" is a copy paste.
>
> J-D
>
> On Sat, Aug 1, 2009 at 9:13 AM, lei wang<nutchmaillist@...> wrote:
> > The hadoop 0.20.X move hadoop-dite.xml to core-site.xml and
> hdfs-site.xml,
> > etc.
> > The tutorial claims in requirement ：This version of HBase will only
> > run on Hadoop
> > 0.20.x <http://hadoop.apache.org/core/releases.html>.
> > But in the “Fully-Distributed Operation” chapter，the tutorial still
> > mentions：
> >
> >   - Add a pointer to your HADOOP_CONF_DIR to CLASSPATH in hbase-env.sh
> >   - Add a copy of hadoop-site.xml to ${HBASE_HOME}/conf, or
> >   - If only a small set of HDFS client configurations, add them to
> >   hbase-site.xml
> >
> >  where is hadoop-site.xml ?
> >
> > On Thu, Jul 30, 2009 at 5:04 PM, Jonathan Gray <jlist@...>
(Continue reading)

Permalink | Reply | 9 comments |
headers
lei wang | 1 Aug 2009 18:15
Picon

Re: Stack: Must hbase0.20 go with zookeeper?

Jean,Thank you.
Now i can start my hbase 0.20 shell and it does work in shell.
But I am still puzzled how to run my own java program to use hbase 0.20 api.
In the 0.19.3 version, i can run the java program with add hbase-* .jar and
hadoop-*.jar to my classpath.
Now i change the jar to 0.20, it cann't work.
The error messges is :
*********************************************************************************************************
[org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper]-[WARN]
java.net.UnknownHostException: isiteam2
    at java.net.Inet4AddressImpl.lookupAllHostAddr(Native Method)
    at java.net.InetAddress$1.lookupAllHostAddr(Unknown Source)
    at java.net.InetAddress.getAddressFromNameService(Unknown Source)
    at java.net.InetAddress.getAllByName0(Unknown Source)
    at java.net.InetAddress.getAllByName(Unknown Source)
    at java.net.InetAddress.getAllByName(Unknown Source)
    at java.net.InetAddress.getByName(Unknown Source)
    at
org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper.setQuorumServers(ZooKeeperWrapper.java:129)
    at
org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper.<init>(ZooKeeperWrapper.java:79)
    at
org.apache.hadoop.hbase.client.HConnectionManager$TableServers.getZooKeeperWrapper(HConnectionManager.java:820)
    at
org.apache.hadoop.hbase.client.HConnectionManager$TableServers.locateRootRegion(HConnectionManager.java:837)
    at
org.apache.hadoop.hbase.client.HConnectionManager$TableServers.locateRegion(HConnectionManager.java:515)
    at
org.apache.hadoop.hbase.client.HConnectionManager$TableServers.locateRegion(HConnectionManager.java:491)
    at
(Continue reading)

Permalink | Reply | 2 comments |
headers
Jean-Daniel Cryans | 1 Aug 2009 18:21
Picon
Favicon

Re: Stack: Must hbase0.20 go with zookeeper?

It seems your client cannot connect to the following ZK servers :
isiteam2, isiteam-desktop  because their hostname is unknown to your
client machine (the name of the exception is quite clear on that).

So make sure you can ping those machines using those hostnames. If you
configured your hbase-site correctly, start-hbase.sh should have
started the ZK servers for you.

You need the same but updated jars plus the zookeeper jar.

J-D

On Sat, Aug 1, 2009 at 12:15 PM, lei wang<nutchmaillist@...> wrote:
> Jean,Thank you.
> Now i can start my hbase 0.20 shell and it does work in shell.
> But I am still puzzled how to run my own java program to use hbase 0.20 api.
> In the 0.19.3 version, i can run the java program with add hbase-* .jar and
> hadoop-*.jar to my classpath.
> Now i change the jar to 0.20, it cann't work.
> The error messges is :
> *********************************************************************************************************
> [org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper]-[WARN]
> java.net.UnknownHostException: isiteam2
>    at java.net.Inet4AddressImpl.lookupAllHostAddr(Native Method)
>    at java.net.InetAddress$1.lookupAllHostAddr(Unknown Source)
>    at java.net.InetAddress.getAddressFromNameService(Unknown Source)
>    at java.net.InetAddress.getAllByName0(Unknown Source)
>    at java.net.InetAddress.getAllByName(Unknown Source)
>    at java.net.InetAddress.getAllByName(Unknown Source)
>    at java.net.InetAddress.getByName(Unknown Source)
(Continue reading)

Permalink | Reply | 1 comment |
headers
lei wang | 1 Aug 2009 23:06
Picon

Re: Stack: Must hbase0.20 go with zookeeper?

OK, thanks, i have made it successful.

On Sun, Aug 2, 2009 at 12:21 AM, Jean-Daniel Cryans <jdcryans@...>wrote:

> It seems your client cannot connect to the following ZK servers :
> isiteam2, isiteam-desktop  because their hostname is unknown to your
> client machine (the name of the exception is quite clear on that).
>
> So make sure you can ping those machines using those hostnames. If you
> configured your hbase-site correctly, start-hbase.sh should have
> started the ZK servers for you.
>
> You need the same but updated jars plus the zookeeper jar.
>
> J-D
>
> On Sat, Aug 1, 2009 at 12:15 PM, lei wang<nutchmaillist@...> wrote:
> > Jean,Thank you.
> > Now i can start my hbase 0.20 shell and it does work in shell.
> > But I am still puzzled how to run my own java program to use hbase 0.20
> api.
> > In the 0.19.3 version, i can run the java program with add hbase-* .jar
> and
> > hadoop-*.jar to my classpath.
> > Now i change the jar to 0.20, it cann't work.
> > The error messges is :
> >
> *********************************************************************************************************
> > [org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper]-[WARN]
> > java.net.UnknownHostException: isiteam2
(Continue reading)

Permalink | Reply | 0 comments |

Gmane