| Age | Commit message (Collapse) | Author |
|---|
|
Always enable VRF strict_mode
|
|
The Linux Kernel needs to be told if IPv6 SR enabled packets whether should be
processed or not. This is done using
/proc/sys/net/conf/<iface>/seg6_* variables:
seg6_enabled - BOOL
Accept or drop SR-enabled IPv6 packets on this interface.
Relevant packets are those with SRH present and DA = local.
0 - disabled (default)
not 0 - enabled
Or the VyOS CLI command:
* set protocols segment-routing interface eth0 srv6
|
|
Enable/Disable VRF strict mode, when net.vrf.strict_mode=0 (default) it is
possible to associate multiple VRF devices to the same table. Conversely, when
net.vrf.strict_mode=1 a table can be associated to a single VRF device.
A VRF table can be used by the VyOS CLI only once (ensured by verify()), this
simply adds an additional Kernel safety net, but a requirement for IPv6 segment
routing headers.
|
|
Why: Smoketests fail as they can not establish IPv6 connection to uvicorn
backend server.
https://github.com/vyos/vyos-1x/pull/2481 added a bunch of new smoketests.
While debugging those failing, it was uncovered, that uvicorn only listens on
IPv4 connections
vyos@vyos# netstat -tulnp | grep 8080
(Not all processes could be identified, non-owned process info
will not be shown, you would have to be root to see it all.)
tcp 0 0 127.0.0.1:8080 0.0.0.0:* LISTEN -
As the CLI already has an option to move the API communication from an IP to a
UNIX domain socket, the best idea is to make this the default way of
communication, as we never directly talk to the API server but rather use the
NGINX reverse proxy.
|
|
|
|
|
|
Add by default sysctl TCP congestion control to BBR.
Default value `cubic` is not optimal.
net.core.default_qdisc=fq
net.ipv4.tcp_congestion_control=bbr
|
|
Replace links to the phabricator site from https://phabricator.vyos.net to
https://vyos.dev
|
|
This sysctl has been removed from kernel 6.0.X onwards but its removal was skipped when upgrading the kernel.
See: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/net/netfilter?id=b118509076b39cc5e616c0680312b5caaca535fe
|
|
The initial implementation from commit ac4e07f9 ("rfs: T4689: Support RFS
(Receive Flow Steering)") always adjusted the global rps_sock_flow_entries
configuration. So if RFS was enabled for one NIC but not the other - it did not
work.
According to the documentation:
RFS is only available if the kconfig symbol CONFIG_RPS is enabled (on by
default for SMP). The functionality remains disabled until explicitly
configured. The number of entries in the global flow table is set through:
/proc/sys/net/core/rps_sock_flow_entries
The number of entries in the per-queue flow table are set through:
/sys/class/net/<dev>/queues/rx-<n>/rps_flow_cnt
Both of these need to be set before RFS is enabled for a receive queue. Values
for both are rounded up to the nearest power of two. The suggested flow count
depends on the expected number of active connections at any given time, which
may be significantly less than the number of open connections. We have found
that a value of 32768 for rps_sock_flow_entries works fairly well on a
moderately loaded server.
This commit sets rps_sock_flow_entries via sysctl on bootup leafing the RFS
configuration to the interface level.
|
|
Directed broadcast is described in rfc1812#section-5.3.5.2 and rfc2644.
By default Linux kernel doesn't forward directed broadcast
packets unless both of `/proc/sys/net/ipv4/conf/all/bc_forwarding`
and `/proc/sys/net/ipv4/conf/$iface/bc_forwarding` are set to 1.
|
|
For more information see:
* https://programmersought.com/article/62242485344/
* https://www.spinics.net/lists/netdev/msg332453.html
* https://github.com/FRRouting/frr/blob/master/doc/user/Useful_Sysctl_Settings.md
|
|
Recommended by FRR best deafults
https://github.com/FRRouting/frr/blob/master/doc/user/Useful_Sysctl_Settings.md
|
|
|
|
|
