1.11 upgrade from 1.9 failed in lab machine

Upgrading a lab environment from 1.9 to 1.11 and all packages yum updated
a-ok but foreman-installer dies on handling passenger (looks like its
setting up the old passenger) and httpd isn't starting. Can someone post
what the 1.11 passenger package and ruby runtime should be or let me know
what may have barfed? (the new RPM repos are there and everything package
wise updated fine)

From the manual:

"Upgrading to Foreman 1.11Preparation

When upgrading across more than one version it is recommended to upgrade to
each intermediate version and follow all upgrade instructions for the
previous versions before updating to 1.11."

I would suggest that the 'recommended' above is changed to 'required' :wink:

··· On Tuesday, March 1, 2016 at 3:52:34 PM UTC-5, Byron Miller wrote:

Upgrading a lab environment from 1.9 to 1.11 and all packages yum updated
a-ok but foreman-installer dies on handling passenger (looks like its
setting up the old passenger) and httpd isn’t starting. Can someone post
what the 1.11 passenger package and ruby runtime should be or let me know
what may have barfed? (the new RPM repos are there and everything package
wise updated fine)

Can you share the error/log/thing you're describing?

At a guess if you're using the installer then it might try installing
ruby193-rubygem-passenger-native which is replaced by
tfm-rubygem-passenger-native.

Add --foreman-passenger-ruby-package=tfm-rubygem-passenger-native to the
installer command to change it.

··· On 01/03/16 20:52, Byron Miller wrote: > Upgrading a lab environment from 1.9 to 1.11 and all packages yum > updated a-ok but foreman-installer dies on handling passenger (looks > like its setting up the old passenger) and httpd isn't starting. Can > someone post what the 1.11 passenger package and ruby runtime should be > or let me know what may have barfed? (the new RPM repos are there and > everything package wise updated fine)


Dominic Cleal
dominic@cleal.org

I will try this and get back, it does look like its passenger related and i
didn't know what name to pass.

Should the ruby version still be the /usr/bin/ruby193-ruby ?

··· On Wednesday, March 2, 2016 at 2:25:05 AM UTC-6, Dominic Cleal wrote: > > On 01/03/16 20:52, Byron Miller wrote: > > Upgrading a lab environment from 1.9 to 1.11 and all packages yum > > updated a-ok but foreman-installer dies on handling passenger (looks > > like its setting up the old passenger) and httpd isn't starting. Can > > someone post what the 1.11 passenger package and ruby runtime should be > > or let me know what may have barfed? (the new RPM repos are there and > > everything package wise updated fine) > > Can you share the error/log/thing you're describing? > > At a guess if you're using the installer then it might try installing > ruby193-rubygem-passenger-native which is replaced by > tfm-rubygem-passenger-native. > > Add --foreman-passenger-ruby-package=tfm-rubygem-passenger-native to the > installer command to change it. > > -- > Dominic Cleal > dom...@cleal.org >

The yum update finished, but on service httpd restart i get:

Web application could not be started

undefined method +&#39; for #&lt;Rails::Paths::Path:0x00000004f79d70&gt; (NoMethodError) /opt/theforeman/tfm/root/usr/share/gems/gems/foreman_abrt-0.0.6/lib/foreman_abrt/engine.rb:18:inblock in <class:Engine>'
/opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/initializable.rb:30:in instance_exec&#39; /opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/initializable.rb:30:inrun'
/opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/initializable.rb:55:in block in run_initializers&#39; /opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:226:inblock in tsort_each'
/opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:348:in block (2 levels) in each_strongly_connected_component&#39; /opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:429:ineach_strongly_connected_component_from'
/opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:347:in block in each_strongly_connected_component&#39; /opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:345:ineach'
/opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:345:in call&#39; /opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:345:ineach_strongly_connected_component'
/opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:224:in tsort_each&#39; /opt/rh/rh-ruby22/root/usr/share/ruby/tsort.rb:203:intsort_each'
/opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/initializable.rb:54:in run_initializers&#39; /opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/application.rb:300:ininitialize!'
/opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/railtie.rb:194:in public_send&#39; /opt/rh/rh-ror41/root/usr/share/gems/gems/railties-4.1.5/lib/rails/railtie.rb:194:inmethod_missing'
/usr/share/foreman/config/environment.rb:5:in &lt;top (required)&gt;&#39; /opt/rh/rh-ruby22/root/usr/share/rubygems/rubygems/core_ext/kernel_require.rb:54:inrequire'
/opt/rh/rh-ruby22/root/usr/share/rubygems/rubygems/core_ext/kernel_require.rb:54:in require&#39; config.ru:3:inblock in <main>'
/opt/rh/rh-ror41/root/usr/share/gems/gems/rack-1.5.2/lib/rack/builder.rb:55:in instance_eval&#39; /opt/rh/rh-ror41/root/usr/share/gems/gems/rack-1.5.2/lib/rack/builder.rb:55:ininitialize'
config.ru:1:in new&#39; config.ru:1:in<main>'
/usr/share/gems/gems/passenger-4.0.18/helper-scripts/rack-preloader.rb:105:in eval&#39; /usr/share/gems/gems/passenger-4.0.18/helper-scripts/rack-preloader.rb:105:inpreload_app'
/usr/share/gems/gems/passenger-4.0.18/helper-scripts/rack-preloader.rb:150:in &lt;module:App&gt;&#39; /usr/share/gems/gems/passenger-4.0.18/helper-scripts/rack-preloader.rb:29:in<module:PhusionPassenger>'
/usr/share/gems/gems/passenger-4.0.18/helper-scripts/rack-preloader.rb:28:in `<main>'

Application root/usr/share/foremanEnvironment (value of RAILS_ENV,
RACK_ENV, WSGI_ENV and PASSENGER_ENV)productionRuby interpreter command

/usr/bin/ruby193-ruby

User and groups

uid=608(foreman) gid=607(foreman) groups=607(foreman),52(puppet)

Also, it seems that the new foreman-installer is trying to reinstall the Centos postgres9.2 but i have postgres9-4 packages installed

so i see:

[root@foreman foreman]# foreman-installer --foreman-passenger-ruby-package=tfm-rubygem-passenger-native
Could not start Service[postgresqld]: Execution of '/usr/bin/systemctl start postgresql' returned 1: Job for postgresql.service failed because the control process exited with error code. See "systemctl status postgresql.service" and "journalctl -xe" for details.
/Stage[main]/Postgresql::Server::Service/Service[postgresqld]/ensure: change from stopped to running failed: Could not start Service[postgresqld]: Execution of '/usr/bin/systemctl start postgresql' returned 1: Job for postgresql.service failed because the control process exited with error code. See "systemctl status postgresql.service" and "journalctl -xe" for details.
Could not start Service[foreman-proxy]: Execution of '/usr/bin/systemctl start foreman-proxy' returned 1: Job for foreman-proxy.service failed because the control process exited with error code. See "systemctl status foreman-proxy.service" and "journalctl -xe" for details.

I have manage db blank, so i would normally run db seed/migrate.

-byron

··· On Wednesday, March 2, 2016 at 2:25:05 AM UTC-6, Dominic Cleal wrote: > > On 01/03/16 20:52, Byron Miller wrote: > > Upgrading a lab environment from 1.9 to 1.11 and all packages yum > > updated a-ok but foreman-installer dies on handling passenger (looks > > like its setting up the old passenger) and httpd isn't starting. Can > > someone post what the 1.11 passenger package and ruby runtime should be > > or let me know what may have barfed? (the new RPM repos are there and > > everything package wise updated fine) > > Can you share the error/log/thing you're describing? > > At a guess if you're using the installer then it might try installing > ruby193-rubygem-passenger-native which is replaced by > tfm-rubygem-passenger-native. > > Add --foreman-passenger-ruby-package=tfm-rubygem-passenger-native to the > installer command to change it. > > -- > Dominic Cleal > dom...@cleal.org >

It should now be /usr/bin/tfm-ruby, but ruby193-ruby is provided as a
symlink for compatibility.

··· On 04/03/16 19:49, Byron Miller wrote: > Should the ruby version still be the /usr/bin/ruby193-ruby ?


Dominic Cleal
dominic@cleal.org

> The yum update finished, but on service httpd restart i get:
>
>
> Web application could not be started
>
> undefined method +&#39; for #&lt;Rails::Paths::Path:0x00000004f79d70&gt; (NoMethodError) &gt; /opt/theforeman/tfm/root/usr/share/gems/gems/foreman_abrt-0.0.6/lib/foreman_abrt/engine.rb:18:inblock in <class:Engine>'

This is a bug in the foreman_abrt plugin, I guess it isn't compatible
with 1.10. I'd suggest filing a bug there
(https://github.com/theforeman/foreman_abrt/issues) and perhaps
uninstall it for the time being.

> Also, it seems that the new foreman-installer is trying to reinstall the Centos postgres9.2 but i have postgres9-4 packages installed
>
>
> so i see:
>
>
> [root@foreman foreman]# foreman-installer
> --foreman-passenger-ruby-package=tfm-rubygem-passenger-native Could
> not start Service[postgresqld]: Execution of '/usr/bin/systemctl
> start postgresql' returned 1: Job for postgresql.service failed
> because the control process exited with error code. See "systemctl
> status postgresql.service" and "journalctl -xe" for details.
> /Stage[main]/Postgresql::Server::Service/Service[postgresqld]/ensure: change
> from stopped to running failed: Could not start
> Service[postgresqld]: Execution of '/usr/bin/systemctl start
> postgresql' returned 1: Job for postgresql.service failed because
> the control process exited with error code. See "systemctl status
> postgresql.service" and "journalctl -xe" for details. Could not
> start Service[foreman-proxy]: Execution of '/usr/bin/systemctl start
> foreman-proxy' returned 1: Job for foreman-proxy.service failed
> because the control process exited with error code. See "systemctl
> status foreman-proxy.service" and "journalctl -xe" for details.
>
>
> I have manage db blank, so i would normally run db seed/migrate.

Perhaps check it's set to false, not blank.

··· On 04/03/16 20:03, Byron Miller wrote:


Dominic Cleal
dominic@cleal.org

Thanks! I'll file a bug so it can be tracked.

-byron

··· On Monday, March 7, 2016 at 2:37:31 AM UTC-6, Dominic Cleal wrote: > > On 04/03/16 20:03, Byron Miller wrote: > > The yum update finished, but on service httpd restart i get: > > > > > > Web application could not be started > > > > undefined method `+' for # > (NoMethodError) > > > /opt/theforeman/tfm/root/usr/share/gems/gems/foreman_abrt-0.0.6/lib/foreman_abrt/engine.rb:18:in > `block in ' > > This is a bug in the foreman_abrt plugin, I guess it isn't compatible > with 1.10. I'd suggest filing a bug there > (https://github.com/theforeman/foreman_abrt/issues) and perhaps > uninstall it for the time being. > > > Also, it seems that the new foreman-installer is trying to reinstall > the Centos postgres9.2 but i have postgres9-4 packages installed > > > > > > so i see: > > > > > > [root@foreman foreman]# foreman-installer > > --foreman-passenger-ruby-package=tfm-rubygem-passenger-native Could > > not start Service[postgresqld]: Execution of '/usr/bin/systemctl > > start postgresql' returned 1: Job for postgresql.service failed > > because the control process exited with error code. See "systemctl > > status postgresql.service" and "journalctl -xe" for details. > > > /Stage[main]/Postgresql::Server::Service/Service[postgresqld]/ensure: > change > > from stopped to running failed: Could not start > > Service[postgresqld]: Execution of '/usr/bin/systemctl start > > postgresql' returned 1: Job for postgresql.service failed because > > the control process exited with error code. See "systemctl status > > postgresql.service" and "journalctl -xe" for details. Could not > > start Service[foreman-proxy]: Execution of '/usr/bin/systemctl start > > foreman-proxy' returned 1: Job for foreman-proxy.service failed > > because the control process exited with error code. See "systemctl > > status foreman-proxy.service" and "journalctl -xe" for details. > > > > > > I have manage db blank, so i would normally run db seed/migrate. > > Perhaps check it's set to false, not blank. > > -- > Dominic Cleal > dom...@cleal.org >