I needed to create this NEW bugzilla issue as a clone of: https://bugzilla.redhat.com/show_bug.cgi?id=1220910 I couldn't clone 1220910 since the comments section is now so long, it's preventing anyone from cloning that issue. Hence, this issue as the clone. Rather than add details/comments from original, please see for details: https://bugzilla.redhat.com/show_bug.cgi?id=1220910
Given all the recent code changes regarding memory utilization, plus the updated Ruby & Rails versions, can we get date for either Master or 5.5?
Versions: Miq master appliance (Built 2015-12-08) manageiq-ovirt-master-201512082000-a918d3d3f6.ova # ruby -v ruby 2.2.3p173 (2015-08-18 revision 51636) [x86_64-linux] # rails -v Rails 4.2.5 Initial Refresh (GC %): S - 10.51%-11.87% M - 11.02%-11.85% L - 13.49%-15.07% XL- 19.94%-20.87% Delta Refresh (GC %): S - 6.06% -14.53% M - 8.65% -13.06% L - 9.16% -11.59% XL- 12.27%-13.62% Basic stackprof data will be attached.
Created attachment 1104698 [details] CLI output from stackprof on 4 initial refreshes per provider and 4 delta refreshes per provider
Time spent in garbage collection has is now about 1/4 or less of the time previously spent in garbage collection. This ticket is being closed as we have met the goal of this ticket (reduction in garbage collection time), additionally efforts to reduce the time spent performing garbage collection will continue as part of regular performance work.
Hi Dennis, How to test this issue from QE prospective? Thanks, Ramesh
Hi Ramesh, Stackprof has been the tool of choice to get the data. Inject stackprof in the environment, add a provider then run the report. Alex's Small/Medium/Large/XLarge environments have be the gold standard for reference. Dennis
Reran on 5.5.2.4: # ruby -v ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-linux] # rails -v Rails 4.2.5.1 Initial Refresh (GC %): S - 10.39%-11.05% M - 10.79%-12.46% L - 14.20%-15.21% XL- 20.67%-21.78% Delta Refresh (GC %): S - 6.11%-10.38% M - 8.62%-13.75% L - 9.93%-12.54% XL- 11.52%-13.15% This is a huge improvement compared to the percentage of gc time I have seen in the 5.3 cfme appliance (original BZ).
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-2016:0159