Description of problem: When issuing command "openstack server list --all-projects", response time is greater than 30 seconds. All other commands run within reasonable time. We verified memcached settings to be ok. DB response time is good. DNS servers response time is good. We tried to see if issue was similar to that seen in BZ https://bugzilla.redhat.com/show_bug.cgi?id=2051631 and https://bugs.launchpad.net/nova/+bug/1964149 , but does not seem to be the case. The case (03430474) contains nova-api strace and matching nova-api logs, sosreports, etc. Version-Release number of selected component (if applicable): RHOSP 16.1 How reproducible: Everytime command "openstack server list --all-projects" is issued. Steps to Reproduce: 1. Can be reproduced on customer system. 2. 3. Actual results: When issuing command "openstack server list --all-projects", response time is greater than 30 seconds. Expected results: Response time of a few seconds at most. Additional info:
FTR https://bugzilla.redhat.com/show_bug.cgi?id=2077043 and https://bugzilla.redhat.com/show_bug.cgi?id=2149605#c38
A fix was merged to downstream RHOSP 16.2 branches and should be released with RHOSP 16.2.5. Since this bug was reported for RHOSP 16.1, RHOSP 16.1 is in Maintenance Support phase and there are strict requirements for fixes we deliver, it doesn't make sense to deliver a change that re-enables DNS support for eventlet with RHOSP 16.1 minor update. This bug is going to be closed as duplicate of bug #2051631. Customer should update to latest RHOSP 16.2 version after a fix for bug #2051631 is released. *** This bug has been marked as a duplicate of bug 2051631 ***