FAQ
I am trying to setup a cluster with 5 hosts.

On the database setup page I leave the option selected to use the
integrated database.
Then I press "test connection"

I receive the error messages for Service Monitor, Activity Monitor and Host
Monitor:
*Unknown host ':7432'. Unable to find it from host 'romeo1.lan'.*

See attached screenshot.



What am I doing wrong?
Any advice is highly appreciated.

Search Discussions

  • Vikas Singh at Jun 11, 2013 at 4:54 pm
    Hi Stefan,

    Please provide the hostname of the CM server in the "Database-Hostname"
    textbox. The entry should look like "hostname:7432" instead of just
    ":7432".

    CM tries to fill it automatically by getting hostname where it is running,
    but that seems to be returning empty for your case.

    Vikas
  • Darren Lo at Jun 11, 2013 at 5:30 pm
    Hi Stefan,

    Also check that your /etc/cloudera-scm-server/db.mgmt.properties looks
    correct. There is a bug where this file is sometimes generated with empty
    host names, usually when you are on a VPN connecting to AWS and start the
    embedded db for the first time. We can't correctly resolve the host name
    when on a VPN sometimes.

    If the db.mgmt.properties file is missing the host names, then fix it by
    adding the fully qualified domain name for your host. Don't use
    "localhost". Reboot your scm server (sudo service clouders-scm-server
    restart) and it should work correctly.

    Thanks,
    Darren

    On Tue, Jun 11, 2013 at 9:54 AM, Vikas Singh wrote:

    Hi Stefan,

    Please provide the hostname of the CM server in the "Database-Hostname"
    textbox. The entry should look like "hostname:7432" instead of just
    ":7432".

    CM tries to fill it automatically by getting hostname where it is running,
    but that seems to be returning empty for your case.

    Vikas


    --
    Thanks,
    Darren
  • Stefan at Jun 11, 2013 at 7:42 pm
    Thank you to all of you for your replies.

    I added the SCM hostname to /etc/cloudera-scm-server/db.mgmt.properties,
    restarted the SCM server and started building the cluster from tzhe
    beginning. It worked.


    problem solved.

    btw, I have no VPN, no AWS. It was a clean install on 5 identical CentOS
    6.4 servers.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupscm-users @
categorieshadoop
postedJun 11, '13 at 10:20a
activeJun 11, '13 at 7:42p
posts4
users3
websitecloudera.com
irc#hadoop

3 users in discussion

Stefan: 2 posts Vikas Singh: 1 post Darren Lo: 1 post

People

Translate

site design / logo © 2022 Grokbase