Saturday, May 21, 2016

Hadoop HDFS Web UI


Now goto http://hadoop-yarn.gvace.com:50070/ we will see namenode info from webpage

Now goto http://hadoop-yarn.gvace.com:50090/ we will see secondarynamenode info from webpage






Started:Sat May 21 11:21:01 EDT 2016
Version:2.6.4, r5082c73637530b0b7e115f9625ed7fac69f937e6
Compiled:2016-02-12T09:45Z by jenkins from (detached from 5082c73)
Cluster ID:CID-4fc1b111-3c85-4e0a-9cd5-e44ec8fe0192
Block Pool ID:BP-796093918-192.168.56.95-1463114304816

Block Pool ID: New in Hadoop2,
HDFS files are stored as blocks in hard drive. These blocks were put into a physical pool to be managed.
Every Namenode has a pool, each pool has an ID

Cluster ID and Block Pool ID are concepts for Namenode Federation.




Logs

IN $HADOOP_HOME/logs

hadoop-yushan-namenode-hadoop-yarn.gvace.com.log 156419 bytes May 21, 2016 12:54:32 PM
hadoop-yushan-namenode-hadoop-yarn.gvace.com.out 4913 bytes May 21, 2016 11:31:45 AM



.log file: by log4j, log most application logs, configured by log4j.properties
.out file: output standard output and standard error logs, does not have a lot


log file name format:
[framework]-[username]-[process name]-[host name].log/outhadoop-yushan-namenode-hadoop-yarn.gvace.com.log


Seeing hadoop-yushan-namenode-hadoop-yarn.gvace.com.out
can also run command: ulimit -a


NameNode Journal Status

Edits of NameNode


NameNode Storage



Startup Progress
SecondaryNameNode combines fsimage and edits periodically, and update fsimage, delete  edits.
Each time it combines, it saves a checkpoint

No comments:

Post a Comment