My definition of extreme load is under extreme stress from multiple tcp flows (100 at 100mbit in this case) as generated by flent tcp_nup test going through fq codel and sch_cake. I will be more clear in the future.
To stress Babel itself I use rtod. On Fri, Sep 21, 2018, 3:57 PM StarBrilliant <[email protected]> wrote: > Hi Juliusz, > > I tested babeld-1.8.3 with around 60 routes on 8 nodes (7 x86_64 and 1 > Rasp Pi) and also confirmed it works. > > On Sat, Sep 22, 2018 at 4:56 AM Dave Taht <[email protected]> wrote: > > I also discussed issues with tim (cc'd) was having on his raspberri pi > > (arm) based version. He seems to have got a much stabler result after > > reverting to 1.7.1. What I observed while watching his version "break" > > was that the metrics would inflate a little bit, 2-3 times, and then > > go unreachable, even with other valid alternate routes present. That > > was on kernel ? and kernel ?. > > I haven't noticed this phenomenon. > I also deployed RTT measurement, and I will continue to watch my network. > > > On Sat, Sep 22, 2018 at 4:56 AM Dave Taht <[email protected]> wrote: > > I tested the babel-1.8.3 release under some extreme loads with 90 > > routes present (30 ipv6 total) > > BTW, I disagree with the word "extreme". > One of my productive environment uses BGP full-table from 3 upstream > ISPs (HE, NTT, IIJ, each with 785k routes). > Babeld simply refused to run on this environment, blocking the whole > network without converging, with 100% CPU utilization. > I am curious about whether babeld can be adapted to run with > full-table. And I will start a new mail thread about this question. > > _______________________________________________ > Babel-users mailing list > [email protected] > https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/babel-users
_______________________________________________ Babel-users mailing list [email protected] https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/babel-users
