This issue should be fixed in https://bugzilla.redhat.com/show_bug.cgi?id=1227008
I'm not able to reproduce this. I'll take a look at the logs again but the code is running so quickly now for me. The only part that would have trouble scaling is the queries to update/insert into MiqQueue. If you can point me at a sample database or something, then I can see if I can reproduce.
https://github.com/ManageIQ/manageiq/pull/14695
New commit detected on ManageIQ/manageiq/master: https://github.com/ManageIQ/manageiq/commit/155ddd6eeec8eae50d2849d742d9c4710593e972 commit 155ddd6eeec8eae50d2849d742d9c4710593e972 Author: Gregg Tanzillo <gtanzill> AuthorDate: Fri Apr 7 15:07:02 2017 -0400 Commit: Gregg Tanzillo <gtanzill> CommitDate: Fri Apr 7 16:39:40 2017 -0400 Fix queueing of historical metrics collection - Set end_time to the end of the current day before queueing historical collection to allow put match and update existing queue item - Set last_perf_capture_on to end time when no metrics were available to prevent constantly queueing the same historical range. https://bugzilla.redhat.com/show_bug.cgi?id=1436034 app/models/metric/ci_mixin/capture.rb | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
Verified on 5.9.0.4.