How to Install Katello 3.2 on CentOS 7.x

14 Responses

  1. lzap says:

    Few suggestions: Use timedatectl to set time or timezone. I also recommend chrony for the NTP.

  2. Tomas says:

    Use –foreman-initial-organization and –foreman-initial-location, unless you enjoy using the Default Organisation 🙂

  3. Yash says:

    I have tried to install katello so many time but it always fails with ssl validation error.

  4. Hi Yash,

    Can you paste the exact error or output here.

  5. Gkasera says:

    Getting dependencies error for following rpms:
    rubygem(rake) >= 0.8.3

    Any idea which repo can provide them

  6. Manny says:

    Hey Can you please help me with this error, I dont need need to connect via a proxy though its asking for proxy url

    # foreman-installer –scenario katello –katello-proxy-url ‘http://localhost’
    Parameter capsule-reverse-proxy invalid: “capsule::params::reverse_proxy” is not a valid boolean
    Parameter capsule-qpid-router invalid: “capsule::params::qpid_router” is not a valid boolean
    Parameter certs-regenerate invalid: “certs::params::regenerate” is not a valid boolean
    Parameter certs-regenerate-ca invalid: “certs::params::regenerate_ca” is not a valid boolean
    Parameter katello-repo-export-dir invalid: “katello::params::repo_export_dir” is not an absolute path

  7. John van Zantvoort says:

    CentOS updated their version 7 to 7.4.1708 and packages like shim have been renamed making the new installer fail. Add the following repos to make the installer work again.

  8. Anidil Rajendran says:

    Thanks for the article Pradeep. I successfully installed on Centos 7.2. It did not work on 7.4 version

  9. Daniel Carrington says:

    I’ve tried installing this several times in several scenarios….AWS EC2 instance, local VM, AWS OpsWorks…and every time, I get as far as running foreman-installer and I always get the same error:

    /usr/bin/wget –no-proxy –timeout=30 –tries=40 –wait=20 –retry-connrefused -qO- ‘http://localhost:8080/candlepin/admin/init’ > /var/log/candlepin/cpinit.log 2>&1 && touch /var/lib/candlepin/cpinit_done returned 8 instead of one of [0]

    What am I missing here? I don’t see others having this error, so I’m confused. I’m following along with the steps, but still getting this error every time.

    Thanks for any help!

  10. James Douglas says:

    Same issue as Daniel Carrington on CentOS 7.5.

  11. Theo says:

    Main problem with SSL on CentOS 7 is that candlepin is apparently old and with the default installation tomcat is not allowed to access postgresql server. This is fixed in newer version of candlepin but that is not what is installed so disabling selinux (setenforce 0) allows install to go through for me. Presumably can audit this to find correct command to use. I also tried to use my own certificates and had host of issues with the SSLCertificateChainFile that is added to apache configuration which makes other parts fail. This might not be part of demo install however so not documenting that other than this might help others pinpoint things to look at. I got through a single pass install by hand editing crt files while install was progressing before apache was restarted but after files were written.

Leave a Reply

Your email address will not be published. Required fields are marked *