[NETCONF-98] Module name conflict limits unrelated functionality Created: 09/Nov/15  Updated: 15/Mar/19  Resolved: 02/Dec/15

Status: Resolved
Project: netconf
Component/s: netconf
Affects Version/s: None
Fix Version/s: None

Type: Bug
Reporter: Vratko Polak Assignee: Unassigned
Resolution: Won't Do Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Operating System: All
Platform: All

External issue ID: 4600

 Description   

Currently on Beryllium there is BGPCEP-306 where conflict on "bgp" model seems to have a consequence of Resconf not becoming available (/restconf/modules keeps returning 503).

Is there a reason to prevent Restconf to function?
Is there a way to allow majority of correct models to work?

Combination of BGPCEP-306 and this is currently causing failure of integration/distribution verify job, which prevents projects to add or edit their integration features.

 Comments   
Comment by Robert Varga [ 10/Nov/15 ]

Moving to netconf project.

Comment by Robert Varga [ 10/Nov/15 ]

RESTCONF specification requires the use of module name as the prefix for namespace resolution in URLs. Therefore modulename-to-namespace mapping has to be unique for URLs to be properly resolved.

I think this needs coordinated effort across all projects to give odl-specific models a structure which will be non-colliding. The OpenBGP model is a standard one – I am afraid we cannot move that one.

Comment by Vratko Polak [ 16/Nov/15 ]

> Therefore modulename-to-namespace mapping has to be unique for URLs to be properly resolved.

This Change Request is about what happens when the mapping is not unique anyway (for example users may be installing their own poorly named modules).

Improper resolution could lead to HTTP status 400 response on affected module name, but it is hardly a reason for status 503 on all restconf requests.

Comment by Ryan Goulding [ 01/Dec/15 ]

This is a property of the protocol and a well known limitation of ODL.

Comment by Vratko Polak [ 01/Dec/15 ]

> a well known limitation of ODL.

Well, the correct response is WONTFIX if this limitation is practically unavoidable,
or set milestone to Boron (or later) if it just low priority.

I do not think this Change Request is INVALID.

Comment by Tony Tkacik [ 02/Dec/15 ]

Ok, changed to WONTFIX, since given current constraints it can not be fixed (supporting both at same time).

Generated at Wed Feb 07 20:14:10 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.