Bug 1495142

Summary:	All the internal hosts should be added to NO_PROXY
Product:	OpenShift Container Platform	Reporter:	Gan Huang <ghuang>
Component:	Installer	Assignee:	Tim Bielawa <tbielawa>
Status:	CLOSED ERRATA	QA Contact:	Gan Huang <ghuang>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	3.7.0	CC:	aos-bugs, jokerman, mmccomas, myllynen
Target Milestone:	---	Keywords:	Regression
Target Release:	3.7.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:	Cause: Facts were lost during some internal refactoring Consequence: The list of names to populate in the NO_PROXY field was empty Fix: The facts have been restored Result: The list of NO_PROXY names is correctly defined	Story Points:	---
Clone Of:		Environment:
Last Closed:	2017-11-28 22:12:28 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 Gan Huang 2017-09-25 09:44:03 UTC

Description of problem:
Only the host itself added to NO_PROXY in /etc/sysconfig/atomic-openshift-master-api, /etc/sysconfig/atomic-openshift-master-controllers, /etc/sysconfig/atomic-openshift-node and /etc/origin/master/master-config.yaml. That result in `oc logs..` failed. We should add all the internal hosts to NO_PROXY at least.

Version-Release number of the following components:
openshift-ansible-3.7.0-0.126.4.git.0.3fc2b9b.el7.noarch.rpm

How reproducible:
always

Steps to Reproduce:
1.Trigger proxy installation
<--snip-->
openshift_https_proxy=http://xxxx.redhat.com:3128
openshift_http_proxy=http://xxxx.redhat.com:3128
<--snip-->


Actual results:
Installation succeed, but only the host itself added to NO_PROXY.

Expected results:
All the internal hosts added to NO_PROXY.

Additional info:

Comment 4 Tim Bielawa 2017-10-17 17:33:30 UTC

I believe I have reproduced this error in a 1m2n cluster. The problem is even worse in that scenario as the nodes will fail to register completely. Just as you described in the OP, the only host entry in the sysconfig proxy settings was for the host I was looking at, along with the boilerplate '.svc...' items.

Trying this now on OCP 3.6

Comment 5 Tim Bielawa 2017-10-19 14:17:38 UTC

I have ran tests and verified this bug for OCP 3.7 when proxy parameters are set. I am working on identifying the source of the regression now.

Comment 6 Gan Huang 2017-10-19 15:09:41 UTC

Tim,

I submitted a PR for the issue, at least it worked for me. But uncertain if that's a good way to fix. Hopefully could help you to identify the issue.

Comment 7 Gan Huang 2017-10-19 15:10:22 UTC

Might be a fix: https://github.com/openshift/openshift-ansible/pull/5569/files

Comment 8 Tim Bielawa 2017-10-19 15:13:59 UTC

Ouch. Since that PR has lingered for almost a month it has gotten out of sync. I'll try fixing it up and running the 3.7 proxy tests again.

Comment 9 Tim Bielawa 2017-10-19 18:55:04 UTC

I've fixed up the rebase problem with the original patch you provided and I have successfully tested the patch. Results are shown in github: https://github.com/openshift/openshift-ansible/pull/5569#issuecomment-338001818

Comment 11 Gan Huang 2017-10-26 09:25:12 UTC

Verified in openshift-ansible-3.7.0-0.178.0.git.0.27a1039.el7.noarch.rpm

Proxy variables added correctly. Thanks Tim.

Comment 15 errata-xmlrpc 2017-11-28 22:12:28 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/RHSA-2017:3188