[OPNFLWPLUG-490] [Lithium redesign] flow not sent to switch if priority is not explicitly set. Created: 05/Jun/15 Updated: 27/Sep/21 Resolved: 11/Jun/15 |
|
| Status: | Resolved |
| Project: | OpenFlowPlugin |
| Component/s: | General |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | ||
| Reporter: | Jamo Luhrsen | Assignee: | Unassigned |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Operating System: All |
||
| External issue ID: | 3603 |
| Description |
|
If a priority is not explicitly set, the default (0x8000 == 32768) should 2015-06-05 15:44:46,121 | ERROR | lt-dispatcher-21 | DataTreeChangeListenerActor | 179 - org.opendaylight.controller.sal-distributed-datastore - 1.2.0.SNAPSHOT | Error notifying listener org.opendaylight.controller.md.sal.binding.impl.BindingDOMDataTreeChangeListenerAdapter@2b426549 example flow xml: |
| Comments |
| Comment by Michal Rehak [ 08/Jun/15 ] |
|
Hi, But models can be touched in Be. Current state is that ofp throws an exception at the point where priority is used. Does this break any tests or applications? Isn't it rational to fix rather those? Because this will be the path ofp will walk in Be? Just asking. If there are strong arguments on doing all validations and default value related stuff in ofp than I am all ear. |
| Comment by Jozef Gloncak [ 08/Jun/15 ] |
|
If the solution with default priority is used then there is patch |
| Comment by Jamo Luhrsen [ 08/Jun/15 ] |
|
I raised this bug because this is something was working one way (using 0x8000) as default priority, then changed so that it wasn't using and throwing the NPE. This is for the Li-redesign plugin. Currently, the He-plugin is taking no priority and using default 0x8000 We should keep the same external behavior for these things if possible. From a previous email exchange, I got the idea that using a default Let me know the decision, because I'd like to get a test in CI specific I do agree that we should be more restrictive to what we accept as a |
| Comment by Jozef Gloncak [ 09/Jun/15 ] |
|
Returning bug back to default assignee |
| Comment by Michal Rehak [ 10/Jun/15 ] |
|
Ok, let's keep this behavior for Li and leave mandatorization for Be. |
| Comment by Jamo Luhrsen [ 10/Jun/15 ] |
|
(In reply to michal rehak from comment #5) Michal, please confirm that we are going to have the he-codebase doing this one way (allowing no priority) and the li-codebase (the redesign) doing it the other way (not allowing priority). Thanks, |
| Comment by Michal Rehak [ 11/Jun/15 ] |
|
Hi Jamo, Any changes to this behavior will be (eventually) done in Be. |