::

failed to restart hbase because of time synchronization

2011/12/08 17:42

Missing time synchronization, scm console cannot start properly. Because of time sync, I installed ntpdate like below;

2011/10/07 - [Programming] - CentOS: time syncronization in local network / setting timezone

However time sync is broken after 2 months. So, to prohibit this problem, I wrote 'crontab' on every server.

0 0 * * * /sbin/ntpdate [ntp server] > /dev/null 2>&1

crontab, hadoop, hbase, Linux, ntpdate, Time Sync, time synchronization

트랙백 0 : 댓글 0

Client configuration

Programming/Hadoop

2011/12/02 12:06

https://ccp.cloudera.com/display/CDHDOC/CDH3+Deployment+on+a+Cluster

1. unset current HADOOP_CONF_DIR if set

2.
$ sudo mv [conf dir path] /etc/hadoop-0.20/

3.
$ sudo update-alternatives --install /etc/hadoop-0.20/conf hadoop-0.20-conf /etc/hadoop-0.20/[conf dir name] [priority]

4. (Not sure) restart your machine (Testing on my machine with reboot, it is applied while on server without reboot, it isn't)

5. To query Hadoop configuration alternatives
$ sudo update-alternatives --display hadoop-0.20-conf

6. When running a test from eclipse, add the "-conf <path to conf dir>" option.

7. Run an example (hadoop must be installed locally)

@$ hadoop jar /usr/lib/hadoop/hadoop-*-examples.jar pi 10 100000000@

Follow the progress of the example on the console, on the cluster's "MapReduce administration page":http://[namenode]:50030/ or using the Job Browser provided by "Hue":http://[namenode]:8088/

configuration

트랙백 0 : 댓글 0