PR: https://github.com/vyos/vyos-1x/pull/899

I ver much like this idea. Certificates can then easily be migrated from device to device, and very easy be referenced in a service.

Sorry, this may be my fault. It seems that I only pay attention to modifying the identifier below and forget the top definition. Sorry.

PR https://github.com/vyos/vyos-1x/pull/898

@Dmitry Is it an actual task? Code was rewritten.

Already fixed with "no_tag_node_value_mangle=True"
https://github.com/vyos/vyos-1x/blob/705eddbc7a2caf09c37ecafb27418a764217975a/python/vyos/config.py#L218

Eigrp in the FRR doesn't work correctly.
The routes still live even if neighbors in a shutdown state.

@Cheeze_It can you re-check it?

There is a link to the existing code for configuration mode, not pr.
So we can to add the op-mode function to re-add/reset with a similar logic. Only thoughts

Not working for me as expected in 1.3.0-rc4
In my current working configuration, the duid is in the /var/lib/dhcpv6/dhcp6c_duid file (29 bytes).

In T3640#96973, @Viacheslav wrote:

I think it will be enough to remove the peer and add again.
@hagbard what do you think?
https://github.com/vyos/vyos-1x/blob/d48dddab0509e562209adfb115b0e691b8e47f54/python/vyos/ifconfig/wireguard.py#L197

Not sure about double quotes, but for example for cloud-init configs, it is necessary to use single quotes.
Ideally, the configuration should look like in show configuration commands

I think it will be enough to remove the peer and add again.
@hagbard what do you think?
https://github.com/vyos/vyos-1x/blob/d48dddab0509e562209adfb115b0e691b8e47f54/python/vyos/ifconfig/wireguard.py#L197

PR https://github.com/vyos/vyos-1x/pull/897
Fix path for swanctl.conf file

In T3640#96937, @hagbard wrote:

Wireguard has no link states on the interface, the ip command just does an 'administrative' up down, which won't start a renegotiation. The policy description (remove peer) needs to be removed from the wg interface and re-added, otherwise you need to wait until wg tries to rekey which will then eventually renegotiate the entire connection.
The removal was as far as I recall part of the original vyos code, so it may have been removed at one point, I haven't looked into the code yet.

For NAT, try setting persistent-keepalive, that is supposed to keep the NAT entry active, even if you have no traffic for the tunnel.

@Harliff Try 1.2.7/1.3 it was fixed with commit https://github.com/vyos/vyos-build/pull/138/files#diff-c7d29a506307d9cf8d86c3cd3f65ca4e4058ea442cacdf9a89d2485b56c7417aR67
T2061

@MaxiM In which exact version was a different behavior?

Wireguard has no link states on the interface, the ip command just does an 'administrative' up down, which won't start a renegotiation. The policy description (remove peer) needs to be removed from the wg interface and re-added, otherwise you need to wait until wg tries to rekey which will then eventually renegotiate the entire connection.
The removal was as far as I recall part of the original vyos code, so it may have been removed at one point, I haven't looked into the code yet.

Actually scratch that. I run a HA pair of VyOS routers via VRRP with a transition script on master/backup, and it looks like when it transitions from backup to master, the commit (at the end of the script) still locks in an endless cycle, combined with some sort of memory leak in keepalived-fifo.py (that doesn't occur if commit-archive via scp is not set up).

Looks fine on boot now.

I've not checked later versions. Maybe it was already fixed on 1.2.7 or 1.3/1.4 ?

Now that the Paramiko and Cryptography versions have been updated, does this problem persist with the newer nightlies? @SrividyaA @FileGo

Done with generate public-key-command. loadkey is deprecated and will be removed in a future version.

In T3640#96876, @c-po wrote:

If your host is behind NAT, could it possibly be that the NAT translation entry expired?

Does the following work:

ip link set dev wg0 down; ip link set dev wg0 up

In T3638#96831, @Viacheslav wrote:

Try to set single quotes.

yes, I am using the following version :

If your host is behind NAT, could it possibly be that the NAT translation entry expired?