Bug 1504635

OK, the [1] (candlepin response timeout increase) _is_ resolution. In particular changes per [2].

[1] https://github.com/Katello/katello/pull/6871
[2] https://github.com/Katello/katello/pull/6871/files#diff-d4ada19367f2e37f2adeafa9de778bcc

Upstream bug assigned to jsherril

Moving this bug to POST for triage into Satellite 6 since the upstream issue http://projects.theforeman.org/issues/20308 has been resolved.

In katello-3.0.0.157 this fails after 5 minutes. So the 2 line fix

RestClient::Resource.new(url,
   :ssl_client_cert => OpenSSL::X509::Certificate.new(client_cert),
   :ssl_client_key => OpenSSL::PKey::RSA.new(client_key),
   :ssl_ca_file => ca_file,
   :verify_ssl => ca_file ? OpenSSL::SSL::VERIFY_PEER : OpenSSL::SSL::VERIFY_NONE,
   :open_timeout => 60*20,   ### added
   :timeout => 60*20         ### added
)

doesn't seem to work. It just pushes the time to fail from around 1 minute to 5 minutes.

Per bcourt, since lst weekend these calls are not further required. So then this should be commented out (together with the owner_upstream_export method being called just from here)?

/opt/theforeman/tfm/root/usr/share/gems/gems/katello-3.0.0.154/app/models/katello/glue/provider.rb :

         if manifest_refresh
            zip_file_path = "/tmp/#{rand}.zip"
            upstream = options[:upstream]
            pre_queue.create(:name => "regenerate upstream certificates for: #{self.organization.name}",
                             :priority => 1, :action => [self, :owner_regenerate_upstream_certificates, upstream],
                             :action_rollback => nil)
            pre_queue.create(:name => "export upstream manifest for owner: #{self.organization.name}",
                             :priority => 2, :action => [self, :owner_upstream_update, upstream, options],
                             :action_rollback => nil)
#            pre_queue.create(:name => "export upstream manifest for owner: #{self.organization.name}",
#                             :priority => 3, :action => [self, :owner_upstream_export, upstream, zip_file_path, options],
#                             :action_rollback => nil)
          else

Stephan, 

60*20 is 20 minutes, so i'm not sure why its not working.

Pavel, yes now that candlepin 2.0 is in production, this isn't required at all, and those lines can be commented out.

Thanks Justin.

Let have this bug as improvement to stop calling the manifest export from subscription.rhn.redhat.com, i.e. implement the https://bugzilla.redhat.com/show_bug.cgi?id=1504635#c6 .

Pavel, Justin: I'm don't understand how https://bugzilla.redhat.com/show_bug.cgi?id=1504635#c6 should work with commenting out calling the export method: as we use the path in the other parts of the orchestration.

(In reply to Ivan Necas from comment #14)
> Pavel, Justin: I'm don't understand how
> https://bugzilla.redhat.com/show_bug.cgi?id=1504635#c6 should work with
> commenting out calling the export method: as we use the path in the other
> parts of the orchestration.

For the particular call flow, the commented-out code helps. If that code is really used by other call flows (I dont know if so or not), then it cant be used as a "permanent" solution but only as a workaround "comment out, refresh manifest, uncomment".

To clarify, for everyone looking for a hotfix.

You need to implement https://bugzilla.redhat.com/show_bug.cgi?id=1504635#c5 first.

https://bugzilla.redhat.com/show_bug.cgi?id=1504635#c6 is the wrong configuration. The code should look like this:

         if manifest_refresh
            zip_file_path = "/tmp/#{rand}.zip"
            upstream = options[:upstream]
            #pre_queue.create(:name => "regenerate upstream certificates for: #{self.organization.name}",
            #                 :priority => 1, :action => [self, :owner_regenerate_upstream_certificates, upstream],
            #                 :action_rollback => nil)
            pre_queue.create(:name => "export upstream manifest for owner: #{self.organization.name}",
                             :priority => 2, :action => [self, :owner_upstream_update, upstream, options],
                             :action_rollback => nil)
            pre_queue.create(:name => "export upstream manifest for owner: #{self.organization.name}",
                             :priority => 3, :action => [self, :owner_upstream_export, upstream, zip_file_path, options],
                             :action_rollback => nil)
          else
            zip_file_path = options[:zip_file_path]
          end

Depending on the Satellite version this must be in /opt/theforeman/tfm/root/usr/share/gems/gems/katello-3.0.0.XXX/app/models/katello/glue/provider.rb

Moving this bug to POST for triage into Satellite 6 since the upstream issue http://projects.theforeman.org/issues/21493 has been resolved.

Verified on Satellite 6.3 snap 27 using the provided manifest, manifest refresh completes successfully.

Clearing needinfo

Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA.
> 
> For information on the advisory, and where to find the updated files, follow the link below.
> 
> If the solution does not work for you, open a new bug report.
> 
> https://access.redhat.com/errata/RHSA-2018:0336