Failed to perform First Run of services. CDH version 5.9 -centos7-virtualbox

[#1]

 Zookeeper-error

Hi There,

I'm building a 3 node cdh v5.9 hadoop cluster on centOs7 guests. At the moment, I'm using cloudera manager to install all components. But I'm running into issue every single time. I've attached failure errors for hdfs & zookeper.

Note, there is passwordless scp between servers, DNS is properly configured and also hosts are properly reachable. I've been trying this past 56 days with no success. The same config  is working fine for apache hadoop. 

Hope I'll get some response before the expiry of cdh trail. 

Thanks,

Mrinal

[#2 · Accepted solution]

Do you have the bigtop-detect-javahome at either location? The error in the one image says it isn't there which may indicate an issue with your parcels or packages install (oh below is for parcels).

/opt/cloudera/parcels/CDH/lib/bigtop-utils/bigtop-detect-javahome
/opt/cloudera/parcels/CDH/bin/bigtop-detect-javahome

[#3 · Accepted solution]

Do you have the bigtop-detect-javahome at either location? The error in the one image says it isn't there which may indicate an issue with your parcels or packages install (oh below is for parcels).

/opt/cloudera/parcels/CDH/lib/bigtop-utils/bigtop-detect-javahome
/opt/cloudera/parcels/CDH/bin/bigtop-detect-javahome

[#4]

Still I see below error. Please help.



​

[root@m bin]# ls -lrt /opt/cloudera/parcels/CDH/bin/bigtop-detect-javahome
lrwxrwxrwx 1 root root 42 Jan 12 04:48
/opt/cloudera/parcels/CDH/bin/bigtop-detect-javahome ->
../lib/bigtop-utils/bigtop-detect-javahome
[root@m bin]# ls -lrt
/opt/cloudera/parcels/CDH/lib/bigtop-utils/bigtop-detect-javahome
-rwxr-xr-x 1 root root 3234 Jan 12 04:48
/opt/cloudera/parcels/CDH/lib/bigtop-utils/bigtop-detect-javahome
[root@m bin]#


User Details:

[root@m bin]# id hdfs
uid=1001(hdfs) gid=1001(hdfs) groups=1001(hdfs),982(hadoop)
[root@m bin]# id hadoop
uid=1003(hadoop) gid=1001(hdfs) groups=1001(hdfs)
[root@m bin]#

hduser has passwordless scp to all hosts.

[#5]

Can you upload the full role logs for one or two of the services failing to start?

[#6 · Accepted solution]

@Piks,

I think I have seen this before.

Check your /opt directory on the zookeeper hosts to make sure that it has "execute" for other.

I reproduced what you see by setting the following:

drwxr-xr--.   3 root root   21 Feb  7 13:50 opt

Try running:

chmod 755

So that it looks like this:

drwxr-xr-x.   3 root root   21 Feb  7 13:50 opt

Try starting services again after that.

Regards,

Ben