> Currently, rule_insert() and rule_delete() ofproto provider APIs do not > > > have return values. There are some possible scenarios where rule insertions > > and deletions can fail at run-time even though the static checks during > > rule_construct() had passed previously. > > > > Some possible scenarios for failure of rule insertions and deletions: > > > > **) Rule insertions can fail dynamically in Hybrid mode (both Openflow and > > > Normal switch functioning coexist) where the CAM space could get suddenly > > > filled up by Normal switch functioning and Openflow gets devoid of > > > available space. > > > > **) Some deployments could have separate independent layers for HW rule > > > insertions/deletions and application layer to interact with OVS. HW layer > > > could face any dynamic issue during rule handling which application could > > > not have predicted/captured in rule-construction phase. > > > > > > This patch is the first step to introduce error reporting for rule > > > insertions/deletions from Client back to OVS. > > > Signed-off-by: Aravind Prasad S <raja....@gmail.com>
>>Thanks for working to improve OVS. >>Please don't triple-space your commit message. >>This commit doesn't actually do anything to handle failures. It only reports them. To be acceptable in OVS, it would have to actually handle them. >>(Also, for error reporting purposes, it's not acceptable to just print a number.) Hi Ben, Thanks for the review. Since the changes to delete the erroneous flow could be a lot more, thought , it would be better to push the changes incrementally. Anyways, thanks again and will get back with the necessary changes. Thanks, S. Aravind Prasad On Sat, Jun 16, 2018 at 9:19 PM, Ben Pfaff <b...@ovn.org> wrote: > On Sat, Jun 16, 2018 at 06:14:38PM +0530, Aravind Prasad wrote: > > Currently, rule_insert() and rule_delete() ofproto provider APIs do not > > > > > > have return values. There are some possible scenarios where rule > insertions > > > > and deletions can fail at run-time even though the static checks during > > > > rule_construct() had passed previously. > > > > > > > > Some possible scenarios for failure of rule insertions and deletions: > > > > > > > > **) Rule insertions can fail dynamically in Hybrid mode (both Openflow > and > > > > > > Normal switch functioning coexist) where the CAM space could get suddenly > > > > > > filled up by Normal switch functioning and Openflow gets devoid of > > > > > > available space. > > > > > > > > **) Some deployments could have separate independent layers for HW rule > > > > > > insertions/deletions and application layer to interact with OVS. HW layer > > > > > > could face any dynamic issue during rule handling which application could > > > > > > not have predicted/captured in rule-construction phase. > > > > > > > > > > > > This patch is the first step to introduce error reporting for rule > > > > > > insertions/deletions from Client back to OVS. > > > > > > Signed-off-by: Aravind Prasad S <raja....@gmail.com> > > Thanks for working to improve OVS. > > Please don't triple-space your commit message. > > This commit doesn't actually do anything to handle failures. It only > reports them. To be acceptable in OVS, it would have to actually handle > them. > > (Also, for error reporting purposes, it's not acceptable to just print a > number.) > > Thanks, > > Ben. > _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev