Bug 1619293

Summary:	[free-int] installer should error if non-openshift-* namespace is configured for logging and priorityClass is going to be set
Product:	OpenShift Container Platform	Reporter:	Justin Pierce <jupierce>
Component:	Installer	Assignee:	ewolinet
Status:	CLOSED ERRATA	QA Contact:	Anping Li <anli>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	3.11.0	CC:	aos-bugs, jcantril, jokerman, mmccomas
Target Milestone:	---
Target Release:	3.11.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:	Cause: If logging was not in a namespace that began with 'openshift-' Fluentd was not able to use the "system-cluster-critical" priority class. Consequence: Fluentd would not be able to start up. Fix: We create a priority class for Cluster Logging and configure Fluentd to use that instead. Result: Fluentd is able to start up, even if not installed to an 'openshift-*' namespace.	Story Points:	---
Clone Of:		Environment:
Last Closed:	2018-10-11 07:25:25 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 Justin Pierce 2018-08-20 14:23:38 UTC

Description of problem:
In v3.11.0-0.16.0, it is possible for a standard upgrade of logging to fail using parameters that would have otherwise worked in 3.10. 

After running an upgrade, fluentd pods will not launch due to the error:
creating: pods "logging-fluentd-" is forbidden: pods with system-cluster-critical priorityClass is not permitted in logging namespace. This is because the fluentd pods have "priorityClassName: system-cluster-critical" set, but do not exist in a privileged openshift-* namespace.  


Version-Release number of selected component (if applicable):
v3.11.0-0.16.0

How reproducible:
100%

Steps to Reproduce:
1. Configure logging to target the 'logging' namespace (anything other than openshift-*)
2. Install logging
3. Observe that fluentd will not start due to this feature of k8s: https://kubernetes.io/docs/concepts/configuration/pod-priority-preemption/  -> https://bugzilla.redhat.com/show_bug.cgi?id=1616171#c5

Actual results:
Fluentd pods fail to start. Logging will not function. 

Expected results:
The installer should fail if settings incompatible are selected for logging. Valid options:
1) openshift-* namespace configured and priorityClassName enabled.
2) non-openshift-* namespace configured, but priorityClassName disabled. 

If the installer is otherwise configured, it should error before it unintentionally cripples logging. 

Additional info:
If the user can prevent priorityClassName from being set by a setting in the inventory, documentation on the impact of this should be provided. 

Description of problem:

Version-Release number of the following components:
rpm -q openshift-ansible
rpm -q ansible
ansible --version

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:
Please include the entire output from the last TASK line through the end of output if an error is generated

Expected results:

Additional info:
Please attach logs from ansible-playbook with the -vvv flag

Comment 1 ewolinet 2018-08-20 22:04:18 UTC

Instead of failing with an error, we will be creating a priority class for fluentd to use instead.

https://github.com/openshift/openshift-ansible/pull/9686

Comment 3 Anping Li 2018-09-10 08:27:57 UTC

The bug have been fix in v3.11.0-0.28.0.0

Comment 5 errata-xmlrpc 2018-10-11 07:25:25 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2018:2652