On Mon, Apr 11, 2022 at 06:10:30AM -0600, Shawn Heisey wrote:
> On 4/10/2022 11:32 PM, Willy Tarreau wrote:
> > Interesting, and not much surprising, given that SSL is handled a bit
> > differently. I suspect we'll see other funny stuff. By the way, if you're
> > receiving this in the second process from the first one and the first one
> > is using HTTP to connect to the second, that would also explain it (in
> > this case the communication between the two could be made over TLS and
> > this rule would not match.
> 
> There should be no proxying between the two processes.  I have 2.4.15
> running the same config it already had, except for a few config lines that
> add the alt-svc header to some specific subdomains.  Then I have 2.6-dev5,
> listening only on udp/443.  I chose port 443 because of your note on
> haproxy.org saying that Chrome doesn't like an alternate port.  If I should
> have proxying between the two processes instead, I will need some details
> about how to set that up.  I deduced that there should be no connection
> between the two because if there is, killing the QUIC-enabled process would
> break things.

Ah OK, In my case I preferred to have a small process dedicated to QUIC
that forwards to the first one so that there's still a single set of
stick-tables, health checks, etc.

> On another system, specifically the one I have in AWS, I upgraded to
> 2.6-dev5 and only have one process running.  It always sets the alt-svc
> header.  That was where I discovered ssl_fc isn't set for quic connections. 
> I had implemented a workaround where I had a duplicate frontend only
> listening on udp/443.  Then a kind soul pointed out that I could instead
> have the second frontend listen on port 80 and have a config that always
> redirects; that is a much more elegant solution.

Ah indeed, if you prefer to use unconditional redirects that's cleaner
and it will even show you cleaner stats that are easier to analyse.

> I checked what Google uses in their alt-svc header for the ma value.  It's
> 30 days.  I adjusted mine to 7200 (2 hours).  After I gain more confidence
> in my http/3 setup, I will increase it.

That still seems long to me for an experiment. On haproxy.org we're set to
1 minute at the moment.

Willy

Reply via email to