Bug 1364583
Summary: | rhel-osp-director: upgrade from 7.3 ->8.0 , br-ex on the controllers was missing ip address and pacemaker was unable to bring up the vip for that network. | ||
---|---|---|---|
Product: | Red Hat OpenStack | Reporter: | Omri Hochman <ohochman> |
Component: | rhosp-director | Assignee: | Dan Sneddon <dsneddon> |
Status: | CLOSED WONTFIX | QA Contact: | Omri Hochman <ohochman> |
Severity: | urgent | Docs Contact: | |
Priority: | urgent | ||
Version: | 8.0 (Liberty) | CC: | bfournie, dbecker, jason.dobies, jcoufal, jschluet, kbasil, markmc, mburns, mcornea, morazi, ohochman, rhel-osp-director-maint, sasha, sclewis, tvignaud |
Target Milestone: | async | ||
Target Release: | 8.0 (Liberty) | ||
Hardware: | x86_64 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | If docs needed, set a value | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2018-03-06 22:18:06 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Omri Hochman
2016-08-05 20:08:56 UTC
I've changed the title : rhel-osp-director: upgrade from 7.3 ->8.0 , br-ex on the controllers was missing ip address and pacemaker was unable to bring up the vip for that network. The actual failure occurred on 7.3 -> 8.0 overcloud upgrade. We need to check if it's not something new that got introduced by using this puddle : http://download.lab.bos.redhat.com/rel-eng/OpenStack/8.0-RHEL-7-director/2016-08-05.3/ Basil, Can you review if this is indeed a RC blocker, or can it wait for a fix for GA (0day) or a later maintenance release. Thanks, Scott workaround (worked for me): Before you begin with the overcloud upgrade: 1. On all overcloud nodes run: systemctl stop NetworkManager systemctl mask NetworkManager 2. On all overcloud nodes run: service network restart 3. On one controller run: pcs cluster stop --all wait until the cluster stops. 4. On one controller run: pcs cluster start --all wait until all the resources are running on all controllers. Note: running "systemctl NetworkManager stop" even on interface(s) with NM_CONTROLLED=no may bring it down due to the issue[1] with NM. 1.https://bugzilla.redhat.com/show_bug.cgi?id=1366348 Note: after applying the workaround from Comment #12 -> the following upgrade from 8.0 to 9.0 failed over (disabled NetworkManager) : https://bugzilla.redhat.com/show_bug.cgi?id=1366392 +1, agree with handling under OSP8 As there is a workaround for the upgrade to 8, this no longer is an issue with other upgrades, and we do not have any fixes planned (coming up on 2 years now) I would like to close this bug. Please respond with any objections. |