Bug 1469585

Summary: [Docs][RFE][Director] Enable config to RHSM proxy in heat-templates
Product: Red Hat OpenStack Reporter: Dan Macpherson <dmacpher>
Component: documentationAssignee: Dan Macpherson <dmacpher>
Status: CLOSED CURRENTRELEASE QA Contact: Charelle Collett <ccollett>
Severity: low Docs Contact:
Priority: low    
Version: 12.0 (Pike)CC: dmacpher, emacchi, lbopf, mburns, srevivo
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2017-12-19 04:20:56 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 1314480    
Bug Blocks:    

Description Dan Macpherson 2017-07-11 14:08:38 UTC
== DECRIPTION ==
Is there an option to force the deployed overcloud image to use a proxy for RHSM contact either via:
- CLI argument to openstack overcloud deploy
- Set variable within rhel-registration heat template
Currently I am modifying the overcloud image using virt-edit and manually configuring /etc/rhsm/rhsm.conf to use the proxy so RHEL registration succeeds during overcloud node boot. This is not a very good method for us due to the automation we are building around Director - we would prefer not to have to crack the overcloud image if the customer is using a proxy server in their environment.

== DOCS IMPACT ==
Requires an update to the Registration chapter in the Advanced Overcloud Guide. Possibly a new example on how to config the RHSM proxy settings

Comment 3 Charelle Collett 2017-10-06 05:07:23 UTC
Looks good to me Dan.

Comment 4 Emilien Macchi 2017-10-19 20:27:08 UTC
Sorry Dan, but our team just figured about a limitation of this feature, which will be considered as a bug, fixed and probably backported as soon as we can.

But for OSP12, it will likely be:

rhel_reg_http_proxy_host is a value of the remote proxy and TripleO doesn't test to reach the host before, so if the proxy isn't reachable, RHSM won't work correctly and probably timeout without an explicit reason.

Our team is working on  a fix but I don't think it will make it on time for OSP12 GA. Therefore, we think this has to be documented for our users.

We'll also make sure GSS knows about this one.

Thanks, please let me know any question.

Needinfo, just to make sure you see this one.