Bug 1083414
| Summary: | ceilometer qpid reconnection delay must be more accurate | ||
|---|---|---|---|
| Product: | Red Hat OpenStack | Reporter: | Flavio Percoco <fpercoco> |
| Component: | openstack-ceilometer | Assignee: | Flavio Percoco <fpercoco> |
| Status: | CLOSED ERRATA | QA Contact: | Shai Revivo <srevivo> |
| Severity: | urgent | Docs Contact: | |
| Priority: | urgent | ||
| Version: | 4.0 | CC: | ajeain, apevec, breeler, eglynn, fdinitto, fpercoco, jruzicka, pbrady, sclewis, scohen, sradvan, yeylon |
| Target Milestone: | z4 | Keywords: | OtherQA, Rebase, ZStream |
| Target Release: | 4.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | openstack-ceilometer-2013.2.3-1.el6ost | Doc Type: | Bug Fix |
| Doc Text: |
Cause: reconnection to the Qpid message broker was delayed by a minimum of 60 seconds.
Consequence: broker reconnect operations from ceilometer agents were not triggered in a timely fashion, so the flow of samples would not resume soon enough once the message broker was restored.
Fix: the minimum reconnection delay has been reduced from 60 seconds to 5 seconds.
Result: after the Qpid broker is restarted, the flow of samples is quickly resumed.
|
Story Points: | --- |
| Clone Of: | 1060689 | Environment: | |
| Last Closed: | 2014-05-29 19:56:24 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Comment 2
Alan Pevec
2014-04-04 11:40:08 UTC
OtherQA Process
===============
0. Install packstack --allinone in the usual way
1. Artificially accelerate the cadence of polling for an individual meter by adding the following entry to /etc/ceilometer/pipeline.yaml:
name: image_pipeline
interval: 10
meters:
- "image.size"
transformers:
publishers:
- rpc://
2. Restart the agent responsible for collecting that accelerated meter:
$ sudo service openstack-ceilometer-central restart
3. Note the 10 second cadence of samples for that meter:
$ ceilometer sample-list -m image.size | awk '/image.size/ {print $(NF-1)}'
4. Stop the qpidd broker:
$ sudo service qpidd stop
5. Note that the flow of samples dries up:
$ while true; do ceilometer sample-list -m image.size | awk '/image.size/ {print $(NF-1)}' | head -1 ; echo '---' ; sleep 1 ; done
6. Restart the qpidd broker:
$ sudo service qpidd stop
7. The flow of sample should resume within 5s:
$ while true; do ceilometer sample-list -m image.size | awk '/image.size/ {print $(NF-1)}' | head -1 ; echo '---' ; sleep 1 ; done
Marking as verified due to the expected rapid resumption of message flow being observed.
Correction to OtherQA steps above, step #6 should read: 6. Restart the qpidd broker: $ sudo service qpidd start 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. http://rhn.redhat.com/errata/RHBA-2014-0577.html |