Description of problem: Pulp calculates the applicability for repo in the batch size of 10 but the profiles of the 10th consumer in the batch will always be skipped. For example, if you have 20 consumers, the 10th and 20th consumers will always get 0 applicability because the calculations are skipped.
below is the offending code: server/managers/consumer/applicability.py if len(profiles_to_process) < batch_size: <========= The last profiles in the batch is no including profiles_to_process.append((repo_id, all_profiles_hash, profiles)) else: batch_regenerate_applicability_task.apply_async( (profiles_to_process,), **{'group_id': task_group_id}) profiles_to_process = []
The Pulp upstream bug status is at POST. Updating the external tracker on this bug.
The Pulp upstream bug priority is at Normal. Updating the external tracker on this bug.
The Pulp upstream bug status is at MODIFIED. Updating the external tracker on this bug.
All upstream Pulp bugs are at MODIFIED+. Moving this bug to POST.
Hello, the upstream PR refers to a regression found and fixed in https://github.com/pulp/pulp/pull/3990 Are both PRs required to backport, in order to properly fix this BZ?
(In reply to Pavel Moravec from comment #8) > Hello, > the upstream PR refers to a regression found and fixed in > > https://github.com/pulp/pulp/pull/3990 > > Are both PRs required to backport, in order to properly fix this BZ? That is the regression of Pulp 2.21.2 but Satellite 6.7.1 is using Pulp 2.21.0.1 so I think it is not affected unless someone is applying a Hotfix or something that has the regression commit. Also see https://bugzilla.redhat.com/show_bug.cgi?id=1847019
The Pulp upstream bug status is at CLOSED - CURRENTRELEASE. Updating the external tracker on this bug.
Steps to test. 1. Spin up and register 10 clients to the satellite. 2. Create custom animal repo and sync it 3. Sync rhel tools repo 4. Create cv and add animal repo and rhel tools repo to it. 5. Publish 6. Create an activation key with cv and lce that it's in from step 4. 7. Subscribe the prod from step 2 in the ak 8. Register 10 hosts to ak. 9. Install a downgraded version of one of the packages (I used kangaroo and/or walrus) 10. Install katello-host-tools 11. Create another custom animal repo, but do not sync it 12. Create another cv and add the new animal repo to it and publish. 13. Create a new ak with the cv and lce that it's in from step 12 14 Subscribe the prod from step 11 to ak. 15. Reregister all 10 hosts to the new ak. 16. Run 'subscription-manager refresh; subscription-manager repos --list' on all 10 hosts 17. Sync the new custom animal repo created in step 11. 18. Publish a new version of the cv created in step 12. 19. Check content hosts page Expected: All 10 hosts should have applicable errata count show up on the page. If you drill into the individual host and go to the errata tab, the applicable errata should show up on the list. Actual All 10 hosts have applicable errata count show up on the page. Drilling into the individual host and going to the errata tab shows the applicable errata. Verified on 6.8.0 snap 15. Note: the workaround to get the the count and applicable errata to show up is to run 'subscription-manager refresh; subscription-manager repos --list' on the hosts.
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 (Important: Satellite 6.8 release), 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-2020:4366