Description of problem: Fresh overcloud deployed with Pike build from (2018-03-19.3) and the overcloud nodes have an extra DNS server before the actual set DNSServers parameter. This introduces DNS time outs on yum transactions but possibly other things are affected by the non-existent DNS server 192.168.122.1. DNSServers parameter was included with the deploy command templates: parameter_defaults: DnsServers: ["10.12.64.161"] [root@overcloud-controller-0 ~]# cat /etc/resolv.conf ; Created by cloud-init on instance boot automatically, do not edit. ; # Generated by NetworkManager nameserver 192.168.122.1 nameserver 10.12.64.161 Deploy command and templates included here https://gist.github.com/akrzos/0ef94734304709c457bc14636a33b021 Version-Release number of selected component (if applicable): Pike (OSP 12) Build: 2018-03-19.3 How reproducible: Always with this build Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
we already have a fix for master/queens, and we'll backport it to pike: https://bugzilla.redhat.com/show_bug.cgi?id=1545842 *** This bug has been marked as a duplicate of bug 1545842 ***
(In reply to Emilien Macchi from comment #1) > we already have a fix for master/queens, and we'll backport it to pike: > https://bugzilla.redhat.com/show_bug.cgi?id=1545842 > > *** This bug has been marked as a duplicate of bug 1545842 *** To be clear though, this is with RHEL 7.4 and not RHEL 7.5. [root@overcloud-controller-0 ~]# cat /etc/resolv.conf ; Created by cloud-init on instance boot automatically, do not edit. ; # Generated by NetworkManager nameserver 192.168.122.1 nameserver 10.12.64.161 [root@overcloud-controller-0 ~]# cat /etc/redhat-release Red Hat Enterprise Linux Server release 7.4 (Maipo) [root@overcloud-controller-0 ~]# cat /etc/rhosp-release Red Hat OpenStack Platform release 12.0 (Pike) [root@overcloud-controller-0 ~]# It seems that the previous bug only applied to Queens with RHEL 7.5, are we sure the same issue and fix will work in the case of Pike with RHEL 7.4?