09-18-2013 06:45 PM
I am trying to start a cluster of 5 nodes through cloudera manager 4.7.1. I am getting the following error on all the nodes.
Detecting Cloudera Manager Server...
I have nodes accessible with passwordless ssh. my user has passwordless sudo permissions. My firewall and SELINUX are disabled in all the nodes. I found a few posts on google groups. But I coudnt find appropriate solutions.
I hope someone her can me suggest me something.
09-27-2013 01:23 PM
Hi Ashwin, it appears as if this agent cannot connect to your CM server qfab-spade.imb.uq.edu.au on port 7182. Can you connect to port 7182 on qfab-spade.imb.uq.edu.au from this agent using the following command?
echo "quit" | nc -v localhost 7182
Also make sure Cloudera Manager Server is listening on port 7182 on qfab-spade.imb.uq.edu.au.
fuser -n tcp 7182
Are you trying to run this cluster on AWS?
10-01-2013 12:28 AM
Thanks for replying back to me. I am trying to create cluster not on AWS
But I have fixed that error. I was behind proxy. I tried export http_proxy to env but it didnt work. So I edited /etc/yum.conf and added my proxy details in that. And now I have got past that error. Now I am running into some warnings in the host inpector. Can we discuss this in the same thread or do I have to start a new one?
10-29-2013 01:28 AM
hi, I have the similar problem, plz help!
The different is my cluster use /etc/hosts to resolve hostname
/etc/hosts contain such content:
And cloudera-manager-server is installed on VM-N2, the nodes use the same username(root) and password to ssh.
But the installation wizard can only install VM-N2 successfully, the other nodes always report error when "Detecting Cloudera Manager Server..".
I make sure
(1) the server is running:
$service cloudera-scm-server status
cloudera-scm-server (pid 8011) is running...
echo "quit" | nc -v localhost 7182
nc: connect to localhost port 7182 (tcp) failed: Connection refused
Connection to localhost 7182 port [tcp/*] succeeded!
HTTP/1.1 400 Bad Request
(3) and I run this script in shell
python -c 'import socket; import sys; s = socket.socket(socket.AF_INET); s.settimeout(5.0); s.connect((sys.argv, int(sys.argv))); s.close();' localhost 7182
it does not echo error
Red Hat Enterprise Linux Server release 6.3 (Santiago)
BEGIN host -t PTR 192.168.0.56
188.8.131.52.in-addr.arpa domain name pointer localhost.
using localhost as scm server hostname
BEGIN which python
BEGIN python -c 'import socket; import sys; s = socket.socket(socket.AF_INET); s.settimeout(5.0); s.connect((sys.argv, int(sys.argv))); s.close();' localhost 7182
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "<string>", line 1, in connect
socket.error: [Errno 111] Connection refused
could not contact scm server at localhost:7182, giving up
waiting for rollback request
10-30-2013 11:42 PM
11-30-2013 08:16 PM
I am new to hadoop and tried to install cloudera manager this time, I got the same error like yours, it is like: could not find scm server etc.
I'd like to know what you did to edit /etc/yum.conf and add a proxy details in that.
can copy and paste that part, so I can modify mine accordingly.
01-07-2014 02:06 PM
I got the same error when I install hadoop manager and I am trying to do the same thing you did to fix this error.
However, I dont know what my 'http_proxy' should be. I am new to Hadoop.
Can you share with me the line you added to your yum.conf file?
thank you very much,
01-08-2014 07:46 AM
@robi this page details how to set up an HTTP proxy in your yum.conf file, however you will need to provide the specific values required by those properties. In other words, the server, port, username and password will all be specific to your company network. You can probably discover what they should be set to by contacting your network admin or looking in your Proxy settings of your web browser to see what your browser is using. HTH.