Hello Otto,
I gave it a try with firefox. according to my subjective tests
I could not spot any differences with various setting.
I've decided to try with some memory benchmarks I could find on github [1]. I
did create a fork [2] with my own test runner to try out your diff. To run it
just do som
This diff is a bit of a cleanup and prepare diff to make the configuration
of BGP MPLS VPNs a bit more flexible. Especially there is the request to
route traffic through multiple mpe(4) interfaces per rdomain. So this is
what this starts to clean up. Remove the static mpe interface in the
kroute sp
so far this only handles printing the UDP hellos, it doesnt try to reach
into a TCP stream and print what it finds in there.
output looks like this:
dlg@cpe0 tcpdump$ sudo ./obj/tcpdump -ni vmx1 port ldp
tcpdump: listening on vmx1, link-type EN10MB
12:26:51.700366 192.168.1.24.646 > 224.0.0.2.
On Thu, Jan 17, 2019 at 03:21:58PM -0700, Theo de Raadt wrote:
> - if (la_hold_total < LA_HOLD_TOTAL && la_hold_total < nmbclust / 64) {
> + if (la_hold_total < nmbclust / 64) {
>
> I have disagreed with claudio about this aspect of the diff.
>
> The refresh attempt is the crucial pro
- if (la_hold_total < LA_HOLD_TOTAL && la_hold_total < nmbclust / 64) {
+ if (la_hold_total < nmbclust / 64) {
I have disagreed with claudio about this aspect of the diff.
The refresh attempt is the crucial problem to fix, because of what it has
done to tcp with slowstart etc.
If ref
When arp entries expire on busy gateways the system will most probably
lose some packets. This comes from a) very small limits of how many
packets can be queued during lookup and b) because we expire entries
before they get refreshed.
This diff fixes these issues for me.
It removes the LA_HOLD_TOT
