Hi,

On Wed, Nov 19, 2025 at 09:19:35AM -0500, Jon Chiappetta via Openvpn-devel 
wrote:
> This is something I noticed recently with potential connection
> interruptions which could occur which the original older code checked for!
> 
> https://github.com/OpenVPN/openvpn/pull/910

Yeah, but, no.

Since our code is single-threaded, there is no time frame between
"unlearn" and "relearn" where a packet could come in and be dropped, due
to "address not known".

So yes, on a PUSH_UPDATE with "the same ifconfig again", we do more work
than we would have to do, but it has no adverse effects - and if you are
worried about that, then do not send PUSH_UPDATE with unchanged configs
(given that the client would do a full tun restart on receiving PUSH_UPDATE,
this has way more impact than on the server).


> There was an update pushed earlier this year which undid some important
> logic in terms of learning/unlearning the ifconfig IP address. The older
> original code would check that the new/old IPs don't match first before
> performing that logic. I believe this was important and correct because it
> could cause connection interruptions during the learn/unlearn process if
> the IP address has stayed the same!

The older code did not do proper unlearning at all, so was obviously not
correct.

gert
-- 
"If was one thing all people took for granted, was conviction that if you 
 feed honest figures into a computer, honest figures come out. Never doubted 
 it myself till I met a computer with a sense of humor."
                             Robert A. Heinlein, The Moon is a Harsh Mistress

Gert Doering - Munich, Germany                             [email protected]

Attachment: signature.asc
Description: PGP signature

_______________________________________________
Openvpn-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openvpn-devel

Reply via email to