Please retest with 5.9.4.
Based onhttps://bugzilla.redhat.com/show_bug.cgi?id=1620246#c3 I'm assigning this ticket to the Performance Team to perform an initial triage.
**Status:** Darn. Nothing is displayed for me. My javascript console displays: 1 error complaining that alertData.resource is null (so it couldn't find the type) (I did translate from obfuscated to source code) https://github.com/ManageIQ/manageiq-ui-classic/blob/master/app/assets/javascripts/services/alerts_center_service.js#L492 The resources are fetched from /api/alerts?expand=resources,alert_actions&attributes=assignee,resource&filter[]=resolved=false&filter[]=or+resolved=nil of the 1,000 records, 33 of the "Container Replicator Successfully Created Pod" alerts point to a defunct ContainerReplicator. alert: {"id"=>"1000000000010", "miq_alert_id"=>"1000000000047", "resource_id"=>"1000000000095", "resource_type"=>"ContainerReplicator", "evaluated_on"=>"2018-05-09T15:24:47Z", "result"=>true, "ems_id"=>"1000000000001", "description"=>"Container Replicator Successfully Created Pod"} MiqAlert: {"id":1000000000010,"description":"VM CPU count was increased","created_on":"2018-04-11T14:35:16.743Z","options":{"notifications":{"email":{"from":"","to":[]}}},"db":"Vm","miq_expression":null,"responds_to_events":"vm_reconfigure","enabled":null,"read_only":true,"hash_expression":{"mode":"internal","options":{"operator":"Increased","hdw_attr":"cpu_total_cores"},"eval_method":"reconfigured_hardware_value"},"severity":null} I will submit a PR to fix the javascript error. **Question:** Tuan, Is it ok if I delete these bad alerts so I can get back to the actual performance issue? Thnx, Keenan
Keena, Go ahead and delete the alerts you need to. I have a clone of this appliance up and running. Tuan
Deleted all alert statuses where the resources no longer existed. I did have quite a few mis-deletes, but there are still 1,1834 alert statuses - which is above 1k alert status for the first page. Now, everything seems to work. Sure, the 1.25 sec to load the tags and 9 sec to load the alerts makes for a slowish page. I think this is a bug and not a performance issue ----- notes: MiqAlertStatus.distinct.pluck(:resource_type).each { |model| model_ids = MiqAlertStatus.where(:resource_type => model).where.not(:resource_id => model.constantize.select(:id)).pluck(:id) ; puts "#{model}: #{model_ids.count} (#{model_ids.first})" ; MiqAlertStatus.destroy_all(:id => model_ids) }
https://github.com/ManageIQ/manageiq-ui-classic/pull/4568
New commit detected on ManageIQ/manageiq-ui-classic/master: https://github.com/ManageIQ/manageiq-ui-classic/commit/5f3b6ac78b5f601cc36a068503c1f276beb56e10 commit 5f3b6ac78b5f601cc36a068503c1f276beb56e10 Author: Keenan Brock <keenan> AuthorDate: Wed Aug 29 18:34:30 2018 -0400 Commit: Keenan Brock <keenan> CommitDate: Wed Aug 29 18:34:30 2018 -0400 Handle Alerts with no resources An alert can point to an invalid resource if the resource is deleted We tend to avoid this, but in one case, Container Replicators can go away https://bugzilla.redhat.com/show_bug.cgi?id=1620246 app/assets/javascripts/services/alerts_center_service.js | 5 +- 1 file changed, 3 insertions(+), 2 deletions(-)
Tuan, thanks for your help. you should be all set
Fixed and verified in 5.10.0.18.20181003162715_dfcff5a