Re: [Ryu-devel] Ryu topology with physical loop-back cables

2015-10-16 Thread Tim Upthegrove
Hi folks, On Wed, Oct 14, 2015 at 6:44 PM, Tim Upthegrove wrote: > > If anyone has any thoughts on this, I'd certainly appreciate it. > I wanted to bump this. Does anyone know why https://github.com/osrg/ryu /blob/v3.25/ryu/topology/switches.py#L778 hides links of switches connected back to the

Re: [Ryu-devel] Ryu topology with physical loop-back cables

2015-10-14 Thread Tim Upthegrove
Hi Daniel, On Wed, Oct 14, 2015 at 5:32 PM, Daniel Herrmann wrote: > > I think that you should not remove both lines completely, but instead > modify the first line and remove the second check, such that the > resulting condition is: > > if not src: > return > > Yes, very good point! I t

Re: [Ryu-devel] Ryu topology with physical loop-back cables

2015-10-14 Thread Daniel Herrmann
Hi Tim! On 14.10.2015 23:02, Tim Upthegrove wrote: > Digging through the code a bit, I found out that is caused by the two > lines of code > at https://github.com/osrg/ryu/blob/v3.25/ryu/topology/switches.py#L778. I > tried removing those two lines, and I found that afterwards I was able > to see

[Ryu-devel] Ryu topology with physical loop-back cables

2015-10-14 Thread Tim Upthegrove
Hello, I have a question about the topology discovery implementation as it relates to physical cables looped back from a switch to itself. I noticed that Ryu's topology discovery implementation was not reporting links made of switches connected back to themselves through the REST API provided by