Note that OpenFlow 1.2 has the notion of a group which can have a failover algorithm. Changes to links in the group still result in a notification to the controller, but can also move traffic to another member of the group without controller intervention.
This in itself is not a solution, but it can be leverage in a solution. -Dan On Tue, May 22, 2012 at 2:43 PM, Kim, Hyojoon <jo...@gatech.edu> wrote: > This is nice! Thanks for the pointers!**** > > ** ** > > I have to dig more into these, but one quick question: Is the alternate > path pre-calculate and pre-installed, or is it recalculated installed on > the fly to switches when link fails?**** > > ** ** > > Thanks,**** > > Joon**** > > ** ** > > *From:* Small, Christopher [mailto:chsm...@indiana.edu] > *Sent:* Tuesday, May 22, 2012 2:37 PM > *To:* Kim, Hyojoon > *Cc:* openflow-discuss@lists.stanford.edu > *Subject:* Re: [openflow-discuss] Failure detection and recovery with > OpenFlow**** > > ** ** > > ** ** > > One piece of software that detects and recovers from failures is the OESS > software, developed by Indiana University and deployed on Internet2 NDDI.* > *** > > ** ** > > This software uses the NoX LLSP discovery but also the OpenFlow Port > Status Async messages to detect failures. OESS can be configured to switch > to alternate path if a link failure is detected. Failovers within a second > even with the latencies on a national footprint.**** > > ** ** > > NDDI **** > > http://www.internet2.edu/network/ose/**** > > ** ** > > OESS Code**** > > http://code.google.com/p/nddi/**** > > ** ** > > ** ** > > Chris Small**** > > InCNTRE **** > > Indiana University**** > > chsm...@indiana.edu**** > > ** ** > > ** ** > > On May 22, 2012, at 5:18 PM, Kim, Hyojoon wrote:**** > > > > **** > > Hi all,**** > > **** > > I am curious whether there is a good solution for *data-plane* link > failure detection and recovery in OpenFlow (or any SDN-based solution).*** > * > > **** > > I am aware of NOX or Floodlight’s LLDP-based topology discovery modules, > which can detect link failures. However, I am looking for an OpenFlow > solution that can not only detect, bus also recover from the failure (e.g., > by rerouting or finding alternate paths) automatically.**** > > **** > > The closest I can find is PortLand’s fault tolerant routing feature, > published in SIGCOMM 2009, which uses a “*fabric manager”* (an OpenFlow > controller) to do the recovery.**** > > **** > > Is anyone aware of some other related work or solutions?**** > > **** > > Thanks,**** > > Joon**** > > **** > > _______________________________________________ > > openflow-discuss mailing list > openflow-discuss@lists.stanford.edu > https://mailman.stanford.edu/mailman/listinfo/openflow-discuss**** > > ** ** > > _______________________________________________ > openflow-discuss mailing list > openflow-discuss@lists.stanford.edu > https://mailman.stanford.edu/mailman/listinfo/openflow-discuss > >
_______________________________________________ openflow-discuss mailing list openflow-discuss@lists.stanford.edu https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
