| Summary: | Reduce Memory usage of MiqScheduleWorker or raise memory threshold | ||
|---|---|---|---|
| Product: | Red Hat CloudForms Management Engine | Reporter: | Alex Krzos <akrzos> |
| Component: | Performance | Assignee: | dmetzger |
| Status: | CLOSED ERRATA | QA Contact: | Dave Johnson <dajohnso> |
| Severity: | high | Docs Contact: | |
| Priority: | high | ||
| Version: | 5.5.0 | CC: | cpelland, jhardy, jprause, obarenbo, thenness |
| Target Milestone: | GA | ||
| Target Release: | 5.6.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | 5.6.0.0 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2016-06-29 15:25:36 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: | |
| Attachments: | |||
Created attachment 1112201 [details]
5.5.0.13-2 MiqScheduleWorker Memory Usage
Created attachment 1112202 [details]
5.4.4.2 MiqScheduleWorker Memory Usage
Created attachment 1112207 [details]
evm.log output from pid 8220 (MiqScheduleWorker)
Reviewing the CSV output of memory usage, the growth spurt in (rss, pss, uss) memory right before the worker exits starts between 16:36:57 and 16:37:07:
TimeStamp RSS
2016-01-05 16:36:37.221739 208.42578125
2016-01-05 16:36:47.231512 208.4921875
2016-01-05 16:36:57.241401 208.5
2016-01-05 16:37:07.251246 219.359375
2016-01-05 16:37:17.261163 230.15234375
2016-01-05 16:37:27.266662 238.265625
2016-01-05 16:37:37.277017 245.796875
2016-01-05 16:37:47.286422 252.53125
2016-01-05 16:37:57.293079 252.53125
*** Bug 1314113 has been marked as a duplicate of this bug. *** 5.5.3.2 has a limit of 300MiB for the MiqScheduleWorker and I have not seen this issue occur again on this build. 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/RHBA-2016:1348 |
Created attachment 1112200 [details] 5.5.2.0 MiqScheduleWorker Memory Usage Description of problem: 5.5.2.0 experiences recycling (exiting and restarting) the MiqScheduleWorker in many of the memory baseline tests. The threshold is set to 250.megabytes in the advanced configuration. 5.5.0.13-2 and 5.4.4.2 also has the MiqScheduleWorker memory threshold set to 250.megabytes however it does not recycle in any of the same scenarios. This has been found in two memory baselining scenarios: Capacity and Utilization and Provisioning. Reviewing historical for 5.5.0.13-2 and 5.4.4.2, the MiqScheduleWorker does not recycle in any of the same scenarios suggesting the additional memory growth was recently introduced. The memory usage does increase between 5.4.4.2 to 5.5.0.13-2 as expected (Move to Ruby 2.2) however not enough to come close to the memory threshold. Version-Release number of selected component (if applicable): 5.5.2.0 How reproducible: ~ >50% of the time (7 out of 13 tests showed this failure) Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: Attaching Memory graphs of the MiqScheduleWorker during Capacity and Utilization memory baselining showing the following versions in the same scenario: 5.4.4.2 5.5.0.13-2 5.5.2.0