Description of problem: Still getting the following error after "force cancel" the paused task that was holding the locks. ----------------------------------------------- Conflicts with tasks: https://satellite.example.com/foreman_tasks/tasks/XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX","errors":["Required lock is already taken by other running tasks.\nPlease inspect their state, fix their errors and resume them. ----------------------------------------------- Steps to Reproduce: 1. Publish a content view and somehow make the task paused with error. Stop Pulp services in the middle of publish might help. 2. After that click "force cancel" the paused task in the Task page. Make sure the task status is set to stopped. 3. Publish the same content view again. Actual results: Fail with task conflict. Expected results: No conflict. Additional info: # Satellite 6.9 /opt/theforeman/tfm/root/usr/share/gems/gems/foreman-tasks-3.0.6/app/models/foreman_tasks/lock.rb ... scope :active, -> { joins(:task).where('foreman_tasks_tasks.state != ?', :stopped) } <======= active scope got delete in 6.10 ... ... # returns a scope of the locks colliding with this one def colliding_locks task_ids = task.self_and_parents.map(&:id) colliding_locks_scope = Lock.active.where(Lock.arel_table[:task_id].not_in(task_ids)) <============ has Lock.active scope colliding_locks_scope = colliding_locks_scope.where(name: name, resource_id: resource_id, resource_type: resource_type) unless exclusive? colliding_locks_scope = colliding_locks_scope.where(:exclusive => true) end colliding_locks_scope end # Satellite 6.10 /opt/theforeman/tfm/root/usr/share/gems/gems/foreman-tasks-4.1.5/app/models/foreman_tasks/lock.rb # returns a scope of the locks colliding with this one def colliding_locks task_ids = task.self_and_parents.map(&:id) colliding_locks_scope = Lock.where(Lock.arel_table[:task_id].not_in(task_ids)) <========== Missing Lock.active scope colliding_locks_scope.where(resource_id: resource_id, resource_type: resource_type) end
Created redmine issue https://projects.theforeman.org/issues/34343 from this bug
Upstream bug assigned to aruzicka
Thanks for the quick fixing. The patch works great for me. It even seems to solve the 'original' problem that created the paused task that after the unlock still held locks. When i try the problematic task to do a promotion of a composite contentviews i still the following once and then task get paused: ~~~ 2022-01-28T16:24:57 [I|bac|ab8bbddb] Task {label: Actions::Katello::ContentView::Publish, id: 14b87c8f-48c0-46eb-a15c-aa65519c86cb, execution_plan_id: 32870ddc-97b4 -40a6-9e78-652cffa1da3f} state changed: running 2022-01-28T16:24:57 [E|bac|ab8bbddb] Error message: the server returns an error ab8bbddb | HTTP status code: 404 ab8bbddb | Response headers: {"Date"=>"Fri, 28 Jan 2022 16:24:57 GMT", "Server"=>"gunicorn", "Content-Type"=>"application/json", "Vary"=>"Accept,Cookie", "Allow"=> "GET, PUT, PATCH, DELETE, HEAD, OPTIONS", "X-Frame-Options"=>"SAMEORIGIN", "Content-Length"=>"23", "Correlation-ID"=>"ab8bbddb-7403-4a61-adef-fcf1c9ad390f", "Access -Control-Expose-Headers"=>"Correlation-ID", "Via"=>"1.1 li-lc-2222.hag.hilti.com"} ab8bbddb | Response body: {"detail":"Not found."} (PulpRpmClient::ApiError) ab8bbddb | /opt/theforeman/tfm/root/usr/share/gems/gems/pulp_rpm_client-3.13.3/lib/pulp_rpm_client/api_client.rb:83:in `call_api' ab8bbddb | /opt/theforeman/tfm/root/usr/share/gems/gems/pulp_rpm_client-3.13.3/lib/pulp_rpm_client/api/repositories_rpm_api.rb:438:in `read_with_http_info' ab8bbddb | /opt/theforeman/tfm/root/usr/share/gems/gems/pulp_rpm_client-3.13.3/lib/pulp_rpm_client/api/repositories_rpm_api.rb:385:in `read' ab8bbddb | /opt/theforeman/tfm/root/usr/share/gems/gems/katello-4.1.1.39/app/lib/actions/pulp3/repository/save_version.rb:45:in `fetch_version_href' ab8bbddb | /opt/theforeman/tfm/root/usr/share/gems/gems/katello-4.1.1.39/app/lib/actions/pulp3/repository/save_version.rb:19:in `run' ~~~ but now with this patch i can cleanly cancel the paused task with the above error. And suprise... after that i can restart the promotion task and it finishes successful.
Moving this bug to POST for triage into Satellite since the upstream issue https://projects.theforeman.org/issues/34343 has been resolved.
Verified on Satellite 6.11 snap 15, after unlocking a failed publish task the cv can be republished successfully.
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 (Moderate: Satellite 6.11 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-2022:5498