On Tue, Nov 12, 2019 at 09:06:31AM +0100, Patrik Rehuš wrote:
Hello,
In my topology I have 2 parallel connections from one host to the same destination. The goal is to switch route to another one under 1 second (in case of broken link). For this I used OSPF with BFD. BFD use 40ms interval and multiplier is 5, so theoretical switching time to another link should be 200ms. In the measurements I find out that sometimes it is OK but sometimes it takes approximately 1 second longer (based on Wireshark it takes approx. 1,2s to switch). I think that it is due to tick parameter in OSPF settings, because base on the documentation „The routing table calculation and clean-up of areas' databases is not performed when a single link state change arrives. To lower the CPU utilization, it's processed later at periodical intervals of num seconds. The default value is 1“.
So, in my case I think that this feature destroys ability of BFD to switch link under 1 second. Is it correct? If yes, is it possible to set lower value than 1 second or completely disable this feature?
Hello No, that is limitation of current implementation. I plan to implement some smarter SPF delay algorithm (e.g. RFC 8405) in the future. -- Elen sila lumenn' omentielvo Ondrej 'Santiago' Zajicek (email: santiago@crfreenet.org) OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net) "To err is human -- to blame it on a computer is even more so."