Created attachment 1422426 [details] iptables rules Description of problem: performed mixed versions stage upgrade, undercloud upgrade to RHOS13 and scaled out 1 extra compute node, the scale out of that node timed out. node did not received dhcp, then the scale out stuck 06:59:49.626034 52:54:00:0f:20:b7 > Broadcast, ethertype IPv4 (0x0800), length 438: 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 52:54:00:0f:20:b7, length 396 Version-Release number of selected component (if applicable): openstack-ironic-api-10.1.2-2.el7ost.noarch openstack-ironic-conductor-10.1.2-2.el7ost.noarch openstack-ironic-common-10.1.2-2.el7ost.noarch instack-undercloud-8.4.0-3.el7ost.noarch How reproducible: 100% Steps to Reproduce: 1. deploy RHOS12 2. upgrade the undercloud 3. perform scale out for additional compute node Actual results: Scale out timeout after 1 hour Expected results: Additional info: var/log/ironic-inspector/ironic-inspector.log:198:2018-04-16 02:58:18.160 27985 DEBUG ironic_inspector.firewall [-] Ignoring failed iptables ('-D', 'INPUT', '-i', 'br-ctlplane', '-p', 'udp', '--dport', '67', '-j', 'ironic-inspector'): iptables v1.4.21: Couldn't load target `ironic-inspector':No such file or directory. . Try `iptables -h' or 'iptables...
Created attachment 1422427 [details] ironic conductor log
Disabling ipatbles does the trick - the ironic introspection pass
This is being tracked in https://bugzilla.redhat.com/show_bug.cgi?id=1570386, patches are upstream. *** This bug has been marked as a duplicate of bug 1570386 ***