Dear Vinithra and Philip ,
Thanks for your concern over the voiced out issue !
Yes Philip , You were right , we kept getting the hearbeat and that was how
the cluster could get the details . But after that we shut down the entire
cluster , manually cleaned the data from the Postgres database and yet we
kept getting the error.
Then finally we recreated the users and groups that cloudera creates with
the linux system manually. On the creation of the users , the issue was
resolved. I'm not sure if this is how the problem could be rectified , but
this provided us with an solution. Previously we got these users and
related groups deleted and thought that on the re installation of the
cluster , these users will be automatically created. But it did not work
that way ,then we created the users and groups manually
PS: Cloudera creates users with the linux system like hadoop , hbase ,
oozie etc ..
Am also pasting the commands of the users that we created:
sudo useradd cloudera-scm
sudo useradd zookeeper
sudo useradd hdfs
sudo useradd hbase
sudo useradd yarn
sudo useradd mapred
sudo useradd hive
sudo useradd oozie
sudo useradd hue
sudo useradd hadoop
sudo groupadd cloudera-scm
sudo groupadd zookeeper
sudo groupadd hdfs
sudo groupadd hbase
sudo groupadd yarn
sudo groupadd mapred
sudo groupadd hive
sudo groupadd oozie
sudo groupadd hue
sudo groupadd hadoop
sudo usermod -a -G cloudera-scm cloudera-scm
sudo usermod -a -G zookeeper zookeeper
sudo usermod -a -G hdfs hdfs
sudo usermod -a -G hbase hbase
sudo usermod -a -G yarn yarn
sudo usermod -a -G mapred mapred
sudo usermod -a -G hive hive
sudo usermod -a -G oozie oozie
sudo usermod -a -G hue hue
sudo usermod -a -G hadoop hadoop
sudo usermod -a -G hadoop hdfs
sudo usermod -a -G hadoop yarn
sudo usermod -a -G hadoop mapred
Thanks all !
Now the cluster is working good :)
On Thursday, October 11, 2012 8:41:20 PM UTC+5:30, manufan wrote:
Hi All .
I installed "Cloudera manager 4.0 free edition" and it was working good
for quite sometime. In between I changed the privileges of certain users
and deleted few users and from then on , the cluster is not working
We tried re installing the entire cluster by deleting all the hosts
related to it , but still we are facing difficulties in installing the
Cluster. The tables consisting the details of the cloudera nodes also was
deleted , yet when the installation is done , it automatically picks up the
older details and shows the respective hosts.
Certain users/Groups were found to be missing in many of the nodes , Not
sure if this might be the reason or Am I approaching it wrong. How to
resolve this issue ?? Is this a known bug ?
Or did the change in permissions alter the setup of the cluster ? Could
the files previously under the Superuser rights , if changed to normal user
will reflect some changes ?
Kindly Help me out on this issue ..