Red Hat Bugzilla – Bug 1404346
[Scale] Full refresh taking a lot of time for RHV provider
Last modified: 2018-03-01 08:08:54 EST
Description of problem: A full refresh takes a lot of time (for ~2200 VMs, it takes 17 minutes for a full refresh, when CFME appliance and RHV reside in the same geographical region). Need to analyze what is taking most of the time, and check how this timing can be improved. Version-Release number of selected component (if applicable): CFME-5.7.0.14 RHV-4.0.5
Please assess the importance of this issue and update the priority accordingly. Somewhere it was missed in the bug triage process. Please refer to https://bugzilla.redhat.com/page.cgi?id=fields.html#priority for a reminder on each priority's definition. If it's something like a tracker bug where it doesn't matter, please set it to Low/Low.
Clearing need info on me, as I am not suppose to set the priority, but the PM.
Ilanit, can you re-test with v4 enabled and with the latest changes from Boris?
Boris's change (Disk Preloading, async part) is yet merged. Therefore, this cannot be tested yet.
Boris, this will be probably improved using the preloading disks + async requests. Please use this bug for the relevant PRs.
https://github.com/ManageIQ/manageiq-providers-ovirt/pull/84
New commit detected on ManageIQ/manageiq-providers-ovirt/master: https://github.com/ManageIQ/manageiq-providers-ovirt/commit/73726e9205146e4f6132960b383e25019e17dfe9 commit 73726e9205146e4f6132960b383e25019e17dfe9 Author: Boris Odnopozov <bodnopoz@redhat.com> AuthorDate: Tue Jul 25 16:32:43 2017 +0300 Commit: Boris Odnopozov <bodnopoz@redhat.com> CommitDate: Tue Oct 3 10:15:23 2017 +0300 Make full refresh through OvirSDK use asynchronous calls to oVirt This should make the refresh significantly faster https://bugzilla.redhat.com/show_bug.cgi?id=1404346 .../redhat/infra_manager/api_integration.rb | 2 + .../redhat/infra_manager/futures_collector.rb | 100 + .../infra_manager/inventory/strategies/v4.rb | 156 +- config/settings.yml | 2 + .../redhat/infra_manager/futures_collector_spec.rb | 103 + .../infra_manager/inventory/strategies/v4_spec.rb | 66 +- .../infra_manager/refresh/refresher_4_1_spec.rb | 32 - .../refresh/refresher_4_async_spec.rb | 655 ++ spec/support/ovirt_sdk/connection_vcr.rb | 46 + .../inventory/strategies/v4/collect_disks.yml | 4046 +++++++++++++ .../refresh/ovirt_sdk_refresh_recording.yml | 6376 ++++++++++++++++++++ 11 files changed, 11490 insertions(+), 94 deletions(-) create mode 100644 app/models/manageiq/providers/redhat/infra_manager/futures_collector.rb create mode 100644 spec/models/manageiq/providers/redhat/infra_manager/futures_collector_spec.rb create mode 100644 spec/models/manageiq/providers/redhat/infra_manager/refresh/refresher_4_async_spec.rb create mode 100644 spec/support/ovirt_sdk/connection_vcr.rb create mode 100644 spec/vcr_cassettes/manageiq/providers/redhat/infra_manager/inventory/strategies/v4/collect_disks.yml create mode 100644 spec/vcr_cassettes/manageiq/providers/redhat/infra_manager/refresh/ovirt_sdk_refresh_recording.yml
Full refresh was measured on CFME-5.9.0.14/RHV-4.2.1 Scale env(1800 VMs, 400 hosts). CFME-5.9 Graph refresh On: 2 min CFME-5.9 Graph refresh Off: 3 min CFME-5.8: 4 min * Initial refresh takes in avg ~1 minute more, on CFME-5.8/5.9, as it includes more operation. Moving bug to verified.
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-2018:0380