Agreed on all counts.
On Wed, May 08, 2013 at 12:31:45PM -0700, Ethan Jackson wrote:
> Yep this all seems reasonable to me, however I don't think it should
> be done specifically for BFD. Instead we should do it for all
> protocols (CFM LACP BFD) which have something to say about the
> liveness o
Yep this all seems reasonable to me, however I don't think it should
be done specifically for BFD. Instead we should do it for all
protocols (CFM LACP BFD) which have something to say about the
liveness of a link. That said, I don't intend to put time into this
at the moment. Feel free to submi
On Wed, May 08, 2013 at 05:48:52AM +, Rajahalme, Jarno (NSN - FI/Espoo)
wrote:
> Ethan,
>
> Looking at the code, it seems to me that there is no link from BFD
> to OpenFlow PORT_STATUS message. I might be wrong, but it seems
> possible that BFD indicated port state changes could be communicat
Ethan,
Looking at the code, it seems to me that there is no link from BFD to OpenFlow
PORT_STATUS message. I might be wrong, but it seems possible that BFD indicated
port state changes could be communicated to the controller with the OFPPS_LIVE
(= OFPUTIL_PS_LIVE) status bit, using the ofproto_
> Since I was mostly happy with the first version, and you addressed
> that stuff, and we're planning to continue working on this, I think
> you can push it without another review. (I'm feeling some time
> pressure lately.)
Sounds good to me. Thanks. I'll merge this now.
Ethan
On Mon, May 06, 2013 at 12:11:24PM -0700, Ethan Jackson wrote:
> Traditionally, Open vSwitch has used a variant of 802.1ag "CFM" for
> interface liveness detection. This has served us well until now,
> but has several serious drawbacks which have steadily become more
> inconvenient. First, the 80
On Mon, May 06, 2013 at 12:10:29PM -0700, Ethan Jackson wrote:
> Since it's been so long, I'll go ahead and resend the latest version
> of the patch. I think it's pretty much reviewed and ready to go, so
> don't feel obligated to comb over it as nothings changed. Feel free
> to make additional co
Traditionally, Open vSwitch has used a variant of 802.1ag "CFM" for
interface liveness detection. This has served us well until now,
but has several serious drawbacks which have steadily become more
inconvenient. First, the 802.1ag standard does not implement
several useful features forcing us to
Sorry to not have responded in so long on this. It's become important
again, so I'm going to try to get it merged soon. Seems pretty close.
> No, that's fine, but I found it hard to infer the units other than
> from the type. Could some comments mention that they are msecs?
I added a comment i
I forgot that I needed to respond here. Sorry, hope late is better
than never.
On Tue, Apr 09, 2013 at 05:40:44PM -0700, Ethan Jackson wrote:
> > The BFD timers are 32-bit counts of microseconds. The timers in
> > struct bfd are 64-bit and appear to be milliseconds, but I'm not
> > certain that
>> +/* XXX Finish Experimnetal BFD support.
>
> What part is experimental? (The protocol or the implementation?)
I consider the implementation experimental until we've tested it more
thoroughly. That said, it's a bit odd to put that in the code, so I've removed
it.
>> + * - CFM random VLAN opti
On Tue, Mar 12, 2013 at 07:44:10PM -0700, Ethan Jackson wrote:
> Traditionally, Open vSwitch has used a variant of 802.1ag "CFM" for
> interface liveness detection. This has served us well until now,
> but has several serious drawbacks which have steadily become more
> inconvenient. First, the 80
Traditionally, Open vSwitch has used a variant of 802.1ag "CFM" for
interface liveness detection. This has served us well until now,
but has several serious drawbacks which have steadily become more
inconvenient. First, the 802.1ag standard does not implement
several useful features forcing us to
13 matches
Mail list logo
