Bug 1734338
| Summary: | Targeted refresh does not occur for openstack | |||
|---|---|---|---|---|
| Product: | Red Hat CloudForms Management Engine | Reporter: | Rahul Chincholkar <rchincho> | |
| Component: | Providers | Assignee: | Marek Aufart <maufart> | |
| Status: | CLOSED EOL | QA Contact: | Ido Ovadia <iovadia> | |
| Severity: | high | Docs Contact: | Red Hat CloudForms Documentation <cloudforms-docs> | |
| Priority: | medium | |||
| Version: | 5.10.6 | CC: | cloudforms-docs, dmetzger, dumindaj, iovadia, jdeubel, jfrey, jhardy, jocarter, maufart, mfeifer, mshriver, niroy, obarenbo, rchincho, simaishi, smallamp | |
| Target Milestone: | GA | Keywords: | Reopened, TestOnly, ZStream | |
| Target Release: | 5.12.0 | Flags: | mfeifer:
mirror+
|
|
| Hardware: | Unspecified | |||
| OS: | Unspecified | |||
| Whiteboard: | ||||
| Fixed In Version: | Doc Type: | If docs needed, set a value | ||
| Doc Text: | Story Points: | --- | ||
| Clone Of: | 1712049 | |||
| : | 1740767 1784107 (view as bug list) | Environment: | ||
| Last Closed: | 2020-03-25 18:03:28 UTC | Type: | --- | |
| Regression: | --- | Mount Type: | --- | |
| Documentation: | --- | CRM: | ||
| Verified Versions: | Category: | Bug | ||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | ||
| Cloudforms Team: | Openstack | Target Upstream Version: | ||
| Embargoed: | ||||
| Bug Depends On: | 1692892, 1712049 | |||
| Bug Blocks: | 1784107, 1740767 | |||
|
Comment 2
CFME Bot
2019-07-30 12:17:11 UTC
New commit detected on ManageIQ/manageiq-providers-openstack/master: https://github.com/ManageIQ/manageiq-providers-openstack/commit/d854ee9c7bf66ec7d22fd14347cbc06c8ff33a72 commit d854ee9c7bf66ec7d22fd14347cbc06c8ff33a72 Author: Marek Aufart <maufart> AuthorDate: Tue Jul 30 07:29:11 2019 -0400 Commit: Marek Aufart <maufart> CommitDate: Tue Jul 30 07:29:11 2019 -0400 Make events_backread switchable OpenStack should expect timestamps in UTC. However some production environments might work in different way. The safest way fix seems to be extend event_backread feature with possibility to turn it off. Event backread can be now turned off by setting :event_backread_seconds settings option to number less than 1, e.g. 0. Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1734338 lib/manageiq/providers/openstack/legacy/events/openstack_ceilometer_event_monitor.rb | 14 +- 1 file changed, 9 insertions(+), 5 deletions(-) New commit detected on ManageIQ/manageiq-providers-openstack/ivanchuk: https://github.com/ManageIQ/manageiq-providers-openstack/commit/3c5e6758acb97de1fcdd9b3043e72c2f2094541d commit 3c5e6758acb97de1fcdd9b3043e72c2f2094541d Author: Ladislav Smola <lsmola> AuthorDate: Thu Aug 8 04:59:31 2019 -0400 Commit: Ladislav Smola <lsmola> CommitDate: Thu Aug 8 04:59:31 2019 -0400 Merge pull request #491 from aufi/events_backread_switch Make events_backread switchable (cherry picked from commit a600714cdcfd7617a69102eae86fdbe1c28178b0) https://bugzilla.redhat.com/show_bug.cgi?id=1734338 lib/manageiq/providers/openstack/legacy/events/openstack_ceilometer_event_monitor.rb | 14 +- 1 file changed, 9 insertions(+), 5 deletions(-) The fix of this BZ is possibility turn off events_readback feature which should fix strange timezones differences between CF and OSP. If user needs switch it off, it is needed set :event_backread_seconds Settings option to 0 (zero). Full settings option path is :ems, :ems_openstack, :event_handling, :event_backread_seconds. Marek, I am not sure if this is a right way to test the PR (https://github.com/ManageIQ/manageiq-providers-openstack/pull/491), However I tried it. 1. Modified /opt/rh/cfme-gemset/bundler/gems/cfme-providers-openstack-df58824a22ec/lib/manageiq/providers/openstack/legacy/events/openstack_ceilometer_event_monitor.rb as mentioned in PR commit. (I took backup of previous one). 2. Restarted evmserverd 3. Checked if the events are being captured with default (event_backread_seconds=15) - The events are not being captured and the following log is observed for querying timestamp: [----] I, [2019-08-15T15:38:42.822177 #13246:af1694] INFO -- : Querying OpenStack for events newer than 2019-08-09 03:00:09 UTC... [----] I, [2019-08-15T15:38:43.526052 #13265:b01e90] INFO -- : Querying OpenStack for events newer than 2019-08-09 03:06:42 UTC... [----] I, [2019-08-15T15:38:43.821547 #13274:adb2a4] INFO -- : Querying OpenStack for events newer than ... 4. Modified event_backread_seconds=0 - The events are not being captured here and the following log is observed for querying timestamp: [----] I, [2019-08-15T15:44:13.781982 #13246:af1694] INFO -- : Querying OpenStack for events newer than 2019-08-09 03:00:09 UTC... [----] I, [2019-08-15T15:44:14.848114 #13265:b01e90] INFO -- : Querying OpenStack for events newer than 2019-08-09 03:06:42 UTC... [----] I, [2019-08-15T15:44:15.658787 #13274:adb2a4] INFO -- : Querying OpenStack for events newer than ... Environment: RHOSP 10 CFME 5.10.6.1 Additionally, in CFME 5.9.9.3 event capture and corresponding targeted refresh still works fine without the mentioned change of `event_backread_seconds=0`. Let me know if you want to discuss my findings over a bluejeans call. Failed QA ========= 5.11.0.19, RHOS 10 I have create an image, FIP, volume and router all events listed on ceilometer but no target refresh triggered more than half an hour I have set event_backread_seconds to '0' but no target refresh triggered Ido, can you provide access for Marek as requested in comment #9? This BZ should go to POST once https://github.com/ManageIQ/manageiq-providers-openstack/pull/516 is backported. |