Nice!
Jean-Daniel Cryans wrote:
Added here http://wiki.apache.org/hadoop/Hbase/Troubleshooting#12
J-D
On Mon, Aug 24, 2009 at 5:20 PM, Patrick Hunt<ph...@apache.org> wrote:
No worries. The details are actually interesting/useful, you might consider
adding to your docs in case another user runs into this.
Patrick
Jean-Daniel Cryans wrote:
Patrick,
Basically, yes. Sorry for the lengthy answer ;)
J-D
On Mon, Aug 24, 2009 at 5:09 PM, Patrick Hunt<ph...@apache.org> wrote:
I see, so an inconsistency then wrt name lookup.
Thanks!
Patrick
Jean-Daniel Cryans wrote:
Well the situation is that HBase now generates the myid files and to
find the id we look in the hbase.zookeeper.quorum configuration that
itself generates a temporary zoo.cfg file. To do that we have to
somehow match the machine's own knowledge of its address with what's
in that list. To find our address we use org.apache.hadoop.net.DNS
with the method getDefaultHost and then we go through the list of
machines defined in the HBase configuration. What comes out of DNS
relies on how the OS is configured or it asks a specified dns server
(if provided).
So, in David's situation, he specified an IP address and DNS returns a
hostname so we don't get a match. The resolution in that case is to
fix the configuration by passing hostnames, to change the OS
configuration, to setup a DNS server or to configure/start zookeeper
by hand. From what I've seen, that stuff is never easier but eh, we
still get you a quorum running in the end :P
J-D
On Mon, Aug 24, 2009 at 4:37 PM, Patrick Hunt<ph...@apache.org> wrote:
Hi Jean-Daniel, not sure I get your response fully. Are you saying that
the
configured ip addr was resolved to a hostname, but that hostname didn't
match the list of ip addresses used when defining the zk quorum
machines?
Is
there a workaround you could suggest for ppl who don't have DNS
available?
Should an Hbase JIRA be created for this -- ie is it something you
consider
should be fixed/improved?
Patrick
Jean-Daniel Cryans wrote:
Oh ok well HBase relies on the DNS class shipped with Hadoop to
determine your address. It will try to use a hostname if possible but
what comes out of there really depends on your OS configuration. In
your case, that means that it resolved a hostname instead of an IP
(which is rare) so you should use it instead.
Also this is HBase-specific, ZK isn't really involved.
J-D
On Mon, Aug 24, 2009 at 3:47 PM, Pythonner<python...@gmail.com> wrote:
I forgot to post that line:
<property>
<name>hbase.zookeeper.quorum</name>
<value>192.168.1.xx</value>
</property>
ok, I'll check the guide shipped with HBase.
On Mon, Aug 24, 2009 at 3:43 PM, Jean-Daniel Cryans
<jdcry...@apache.org>wrote:
David,
hbase.master is deprecated in HBase 0.20, instead you have to
specify
hbase.zookeeper.quorum if you want to use HBase in a distributed
mode
with a ZK quorum. Please see the Getting Started documentation
shipped
with HBase.
J-D
On Mon, Aug 24, 2009 at 3:39 PM, Pythonner<python...@gmail.com>
wrote:
Hello,
this is a follow-up of discussion started on twitter with
http://twitter.com/phunt.
I installed HBase 0.20.0 RC2 on Ubuntu server boxes.
If I'm using machines IP in config files (see below), I get the
following
error message:
'Could not find my address: xyz in list of ZooKeeper quorum
servers'
message
(where 'yxz' is a hostname)
my config is:
hbase-env.sh:
export HBASE_MANAGES_ZK=true
hbase-site.xml:
<configuration>
<property>
<name>hbase.rootdir</name>
<value>hdfs://192.168.1.xx:9200/hbase</value>
</property>
<property>
<name>hbase.master</name>
<value>192.168.1.xx:60000</value>
</property>
<property>
<name>hbase.cluster.distributed</name>
<value>true</value>
</property>
</configuration>
from vanilla Ubuntu server install, I removed the 127.0.1.1 line
from
/etc/hosts
Is it supposed to work well with IP addresses only?
David
--
Balie - Baseline Information Extraction
http://balie.sourceforge.net
[Open Source ~ 100% Java ~ Using Weka ~ Multilingual]
