On Thu, Jun 30, 2016 at 04:14:04PM -0400, Russell Bryant wrote: > Prior to this commit, once a connection had been committed to the > connection tracker, the connection would continue to be allowed, even > if the policy defined in the ACL table changed. This patch changes > the implementation so that existing connections are affected by policy > changes. > > The implementation is based on the suggested approach in this mailing > list thread: > > http://openvswitch.org/pipermail/dev/2016-February/065716.html > > The implementation is covered in much more detail in the commit message > for patch 3, as well as code comments and doc updates. > > v1->v2: > - Address issue pointed out by Han Zhou where removing and then re-creating > an ACL did not allow an established connection to continue. The changes > are in patch 3. > v2->v3: > - rebase and resolve conflicts with master. > - Use ct_label instead of ct_mark. > - patch 1: add ACK from han, otherwise unchanged > - patch 2: add support for setting ct_label. v2 only included ct_mark. > I did not include Han's ACK here because the changes were non trivial. > - patch 3: add ACK from han. The rest of the changes are trivial > replacement of ct_mark with ct_label. > v3->v4: > - Added tests for additions to the ct_commit() logical flow action. > - Simplified ct_commit() logical flow action additions as suggested by Ben. > - Lots of doc cleanup as suggested by Justin. > v4->v5: > - Rebase. > - Support a mask for the value of ct_mark or ct_label in the ct_commit() > action. > - Update ovn-northd to explicitly specify that it is only setting 1 bit > of ct_label. > - This version now has all the changes requested by Justin Pettit, so is > ready for his review. > v5->v6: > - Applied patch 1/2 in v5 with minor updates. > - Rebase final patch.
This seems to have multiple acks, do you want particular review of some part of it? _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
