[GBP-20] MatchBuilder construction by using classifier instances pipeline. Created: 20/Mar/15 Updated: 29/May/15 Resolved: 29/May/15 |
|
| Status: | Resolved |
| Project: | groupbasedpolicy |
| Component/s: | General |
| Affects Version/s: | unspecified |
| Fix Version/s: | None |
| Type: | Bug | ||
| Reporter: | Tomas Cechvala | Assignee: | Tomas Cechvala |
| 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: | 2883 |
| Description |
|
The classifier instances pipeline (L4Classifier->EtherTypeClassifier) for MatchBuilder construction does not work sufficiently. Unnecessary constraints limit functionality. E.g. it is not possible to classify traffic based on L4 ports just for ipv4 or ipv6 packet. In fact, current implementation does not take ip version into account at all. |
| Comments |
| Comment by Thomas Bachman [ 03/Apr/15 ] |
|
I think the intent was to have the classifier definitions provide more specificity. If there's a need for an IPv4 only L4 classifier, then you would create a new classifier definition for that (note: to create that specific case, you'd also need to create an IPv4 only classifier definition, and the new L4 classifier definition would call out the IPv4 only classifier definition). That's my understanding of how the current model/definitions are supposed to work. |
| Comment by Keith Burns [ 26/May/15 ] |
|
Are we saying this is not need, Thomas/Tomas ? |
| Comment by Tomas Cechvala [ 27/May/15 ] |
|
It was already merged. https://git.opendaylight.org/gerrit/#/c/17204/ |