[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Errors trying to apply puppet manifest



Can't guide on all but I'll try:

----- Original Message -----
> From: "Jose G N Alban" <jgalbandebritto1 sheffield ac uk>
> To: "Troy Dawson" <tdawson redhat com>
> Cc: users lists openshift redhat com
> Sent: Monday, July 15, 2013 1:12:46 PM
> Subject: Errors trying to apply puppet manifest
> 
> Hello all,
> 
> My environment is:
> - Scientific Linux EL 6
> - puppet 3.2.2
> - ruby193 collection installed
> - SELinux disabled
> - Modified openshift origin puppet module files (console.pp &
> broker.pp) to prevent checking gems dependencies
> - puppet Manifest available on
> http://openshift.github.io/origin/file.install_origin_using_puppet.html#Installing_Puppet
> ("Configuring an all-in-one host")
> 
> After trying to apply the manifest, it lasted ~6900 seconds, and a
> bunch of errors came on. Not sure if it's alright or if I should do
> something to fix it up.
> 
> )1st error raised:
> 
> >> Error:
> >> /Stage[main]/Openshift_origin::Selinux/Selboolean[allow_polyinstantiation]:
> >> Could not evaluate: Execution of '/usr/sbin/getsebool
> >> allow_polyinstantiation' returned 1: /usr/sbin/getsebool:
> >>  SELinux is disabled
> -----
> and it was really disabled:
> 
> > sestatus
> >> SELinux status:                 disabled
> 
> I suppose this error is because the manifest is trying to test access
> permissions, but it can't, right? Is it recommended to enable
> SELinux after all, for Openshift Origin?

You need SELinux enabled to install. It can be in permissive mode (setenforce 0) if you're debugging something, but it must be enabled.

> )2nd error:
> 
> >> Error: /usr/sbin/oo-mongo-setup returned 1 instead of one of [0]
> >> Error:
> >> /Stage[main]/Openshift_origin::Mongo/Exec[/usr/sbin/oo-mongo-setup]/returns:
> >> change from notrun to 0 failed: /usr/sbin/oo-mongo-setup returned
> >> 1 instead of one of [0]
> -----
> I don't understand that one? "returned 1 instead of one of [0]"

The command was supposed to have a return code = 0 (i.e., in the set including 0) but instead returned 1.

> )3rd error (and repeated 20 times):
> 
> >>Error: Execution of '/usr/bin/yum -d 0 -e 0 -y install
> >>openshift-origin-cartridge-mock-plugin' returned 1: Error:
> >>Package:
> >>rubygem-openshift-origin-node-1.10.4-1.git.17.d1335eb.el6.noarch
> >>(openshift-origin)
>            Requires: pam_openshift
>  You could try using --skip-broken to work around the problem
>  You could try running: rpm -Va --nofiles --nodigest

missing the pam_openshift package... not sure why that would be, perhaps the repository is missing it.

> -----
> After some more errors, I've found that other error and seemed
> related:
> 
> )4th error:
> >>Error: Execution of '/usr/bin/yum -d 0 -e 0 -y install
> >>pam_openshift' returned 1: Error: Nothing to do
> >>Error:
> >>/Stage[main]/Openshift_origin::Node/Package[pam_openshift]/ensure:
> >>change from absent to present failed: Execution of '/usr/bin/yum
> >>-d 0 -e 0 -y install pam_openshift' returned 1: Error: Nothing to
> >>do
> -----
> Is it a problem with repositories?

Likely. Either your setup to refer to the repos, or the repos themselves.

> )5th error:
> >> Error: Execution of '/usr/bin/yum -d 0 -e 0 -y install rhc'
> >> returned 1: Error: Package: rhc-1.10.4-1.git.2.4b8658c.el6.noarch
> >> (openshift-origin)
>            Requires: rubygem-test-unit
>  You could try using --skip-broken to work around the problem
>  You could try running: rpm -Va --nofiles --nodigest
> -----
> missing dependency / manifest needs to require it?

Similar as above

> )6th error:
> >> Error: Execution of '/usr/bin/yum -d 0 -e 0 -y install
> >> openshift-origin-msg-node-mcollective' returned 1: Error:
> >> Package: ruby193-mcollective-common-2.2.1-1.el6_3.noarch
> >> (openshift-origin-deps)
>            Requires: mcollective-common = 2.2.1
>            Installed: mcollective-common-2.2.4-1.el6.noarch
>            (@puppetlabs-products)
>                mcollective-common = 2.2.4-1.el6
>            Available: mcollective-common-1.2.1-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 1.2.1-1.el6
>            Available: mcollective-common-2.0.0-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 2.0.0-1.el6
>            Available: mcollective-common-2.0.0-6.el6.noarch (epel)
>                mcollective-common = 2.0.0-6.el6
>            Available: mcollective-common-2.0.1-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 2.0.1-1.el6
>            Available: mcollective-common-2.2.0-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 2.2.0-1.el6
>            Available: mcollective-common-2.2.1-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 2.2.1-1.el6
>            Available: mcollective-common-2.2.1-1.el6_3.noarch
>            (openshift-origin-deps)
>                mcollective-common = 2.2.1-1.el6_3
>            Available: mcollective-common-2.2.2-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 2.2.2-1.el6
>            Available: mcollective-common-2.2.3-1.el6.noarch
>            (puppetlabs-products)
>                mcollective-common = 2.2.3-1.el6
> ----
> It seems that it was available, I'm not sure why it couldn't use it?

The version required is lower than what you have installed. yum won't downgrade what's already installed. Might want to consider doing it without puppetlabs-products; either that, or the specific requirement for mcollective 2.2.1 should be relaxed.

> )7th error:
> >> Error: /usr/sbin/oo-register-dns -h cloud -n 192.168.1.9 returned
> >> 1 instead of one of [0]
> >> Error: /Stage[main]/Openshift_origin::Named/Exec[register broker
> >> DNS]/returns: change from notrun to 0 failed:
> >> /usr/sbin/oo-register-dns -h cloud -n 192.168.1.9 returned 1
> >> instead of one of [0]
> -----
> Maybe it's happening because I'm trying to install with a LAN IP?

It tried to make a DNS update, which is done based on DNS plugin settings. Doesn't say why it failed; perhaps the DNS server is not accepting updates with the key given, or not there at all.

> )8th error:
> >> Notice: /Stage[main]/Openshift_origin::Console/Exec[Console gem
> >> dependencies]/returns: Could not find gem
> >> 'openshift-origin-console (>= 0) ruby' in the gems available
> >> Notice: /Stage[main]/Openshift_origin::Console/Exec[Console gem
> >> dependencies]/returns: on this machine.
> >> Error: /Stage[main]/Openshift_origin::Console/Exec[Console gem
> >> dependencies]: Failed to call refresh: /bin/rm -f Gemfile.lock &&
> >>     /usr/bin/scl enable ruby193 "bundle show" &&     /bin/chown
> >> apache:apache Gemfile.lock &&     /bin/rm -rf tmp/cache/* &&
> >>     /usr/bin/scl enable ruby193 "rake assets:precompile" &&
> >>     /bin/chown -R apache:apache /var/www/openshift/console
> >> returned 7 instead of one of [0]
> >> Error: /Stage[main]/Openshift_origin::Console/Exec[Console gem
> >> dependencies]: /bin/rm -f Gemfile.lock &&     /usr/bin/scl enable
> >> ruby193 "bundle show" &&     /bin/chown apache:apache
> >> Gemfile.lock &&     /bin/rm -rf tmp/cache/* &&     /usr/bin/scl
> >> enable ruby193 "rake assets:precompile" &&     /bin/chown -R
> >> apache:apache /var/www/openshift/console returned 7 instead of
> >> one of [0]
> >> Notice:
> >> /Stage[main]/Openshift_origin::Console/File[/var/www/openshift/console/Gemfile.lock]/ensure:
> >> created
> -----
> Not sure of what's going on here? Maybe some previous problem missing
> "pam_openshift" dependency…

No, looks like a different dependency issue...

> )9th error:
> >>Error: Execution of '/usr/bin/yum -d 0 -e 0 -y install
> >>openshift-origin-cartridge-perl' returned 1: Error: Package:
> >>openshift-origin-cartridge-perl-0.5.3-1.git.13.d1335eb.el6.noarch
> >>(openshift-origin)
>            Requires: perl-MongoDB

Another dependency...

> >>Error: Package:
> >>rubygem-openshift-origin-node-1.10.4-1.git.17.d1335eb.el6.noarch
> >>(openshift-origin)
>            Requires: pam_openshift

Same one as before.

>  You could try using --skip-broken to work around the problem
>  You could try running: rpm -Va --nofiles --nodigest
> -----
> Another dependency not resolved. Repository problems or missing
> puppet manifest declaration or something else?
> ===
> Sorry for the massive amount of information, and I'll be glad to
> provide more information if needed.
> 
> Regards,
> --ja
> 
> _______________________________________________
> users mailing list
> users lists openshift redhat com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
> 


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]