Redis THT templates contain malformed metadata_settings
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
tripleo |
Fix Released
|
High
|
Harry Rybacki | ||
Queens |
Fix Released
|
High
|
Harry Rybacki | ||
Rocky |
Fix Released
|
High
|
Harry Rybacki |
Bug Description
Description
===========
Malformed THT templates (metadata_settings specifically) for Redis are resulting in service principals not being created by noavjoin service. As a result, when during Step2 of deployment the `getcert` request fails on a permission fail.
Steps to reproduce
==================
1. Deploy non-HA undercloud with queens or rocky bits using FreeIPA as your CA.
2. Attempt to deploy overcloud with internal TLS via TripleO e.g.:
openstack overcloud deploy \
--templates \
-e /usr/share/
-e /usr/share/
-e /usr/share/
-e /home/stack/
-e /home/stack/
Expected result
===============
Novajoin adds service principal for Redis to FreeIPA. Overcloud deploys successfully.
Actual result
=============
Deployment blows up during Step2 when `getcert request` is invoked to fetch a certifcate for Redis because it lacks permissions (service principal for Redis was not added to IdM).
Environment
===========
1. Found bug in Queens but verified it exists in Rocky. The issue was resolved during an architectural shift between Rocky and Stein so it does not effect releases beyond Rocky.
2. Which storage type did you use?
Default storage
3. I used FreeIPA as my CA but this should reproduce with other CAs.
Logs and Configs
================
## Overcloud deploy invocation ##
openstack overcloud deploy \
--templates \
--ntp-server clock1.
-e /usr/share/
-e /usr/share/
-e /usr/share/
-e /home/stack/
-e /home/stack/
## cloud-names.yaml ##
parameter_defaults:
CloudDomain: ooo.test
CloudName: overcloud.ooo.test
CloudNameInte
CloudNameStorage: overcloud.
CloudNameStor
CloudNameCtlp
## misc-bits.yaml ##
parameter_defaults:
DnsServers: ["192.168.1.12"] # <-- FreeIPA server
## Deployment log ##
2019-08-01 18:11:32Z [overcloud-
2019-08-01 18:11:33Z [overcloud-
2019-08-01 18:11:33Z [overcloud-
2019-08-01 18:11:33Z [overcloud-
2019-08-01 18:11:33Z [overcloud-
2019-08-01 18:11:33Z [AllNodesDeploy
2019-08-01 18:11:34Z [overcloud]: UPDATE_FAILED Resource UPDATE failed: Error: resources.
Stack overcloud UPDATE_FAILED
## Controller journalctl ##
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Jul 29 17:41:23 overcloud-
Changed in tripleo: | |
status: | Triaged → Fix Released |
Changed in tripleo: | |
status: | Fix Released → In Progress |
Changed in tripleo: | |
milestone: | train-3 → ussuri-1 |
Changed in tripleo: | |
milestone: | ussuri-1 → ussuri-2 |
Changed in tripleo: | |
milestone: | ussuri-2 → ussuri-3 |
Changed in tripleo: | |
milestone: | ussuri-3 → ussuri-rc3 |
Changed in tripleo: | |
status: | In Progress → Fix Released |
Assigning self -- working on fix for Queens/Rocky presently. As noted in the description the issue was fixed during a big code shift from Rocky-->Stein.