Hi, > This is pretty important wrt dual-stack configurations because they can > be implemented with just slight changes to spice server (unlike the old > RFE requesting listening on multiple addresses): > > * when no addr= or ipvx options are set, listen on ::0 spice-server sets "ai.ai_flags = AI_PASSIVE | AI_ADDRCONFIG", which should make getaddrinfo() pick something reasonable, specifically listen on ipv6 only if the machine actually has ipv6 connectivity.

I think this is the behavior we want here. > * when ipv4 and no addr= option is set, listen on 0.0.0.0 Works today. > * when ipv6 is set, set IPV6_ONLY to 1 to make sure that spice server > won't listen on ipv4 Trivially to add, see patch. > * when conflicting ipvx and addr= options are set, error out (this > already works fine) Works today indeed. cheers, Gerd _______________________________________________ Spice-devel mailing list Spice-devel(a)lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/spice-devel

Hi, I learned few things about ipv6 lately. Most importantly about dual-socket that means that a process that opens ::0 automatically listens on both ipv4 _and_ ipv6 unless it sets IPV6_ONLY option of setsockopt() to 0. This is pretty important wrt dual-stack configurations because they can be implemented with just slight changes to spice server (unlike the old RFE requesting listening on multiple addresses): * when no addr= or ipvx options are set, listen on ::0 * when ipv4 and no addr= option is set, listen on 0.0.0.0 * when ipv6 is set, set IPV6_ONLY to 1 to make sure that spice server won't listen on ipv4 * when conflicting ipvx and addr= options are set, error out (this already works fine)

This will affect upper layers though, given bugs like https://bugzilla.redhat.com/show_bug.cgi?id=832121 , but it seems to me like the step in the right direction. Any thoughts/comments before I file this as a bug?

David

Hi, > * new spice-server feature: add option to bind to a selected interface > regardless of its addresses How does that work?

On 06/15/12 12:00, Daniel P. Berrange wrote: > On Fri, Jun 15, 2012 at 11:48:28AM +0200, Gerd Hoffmann wrote: >> Hi, >> >>> * new spice-server feature: add option to bind to a selected interface >>> regardless of its addresses >> >> How does that work? > > I presume the client app would request listen=eth0, and the QEMU > would have to call getifaddrs() to determine what IP addresses > currently correspond to eth0. Ah, so there isn't a direct way I'm not aware of, you still bind to a specific ip address (or multiple), just specified in a different way ;) Note that supporting this isn't going to work with a single listening socket. Having ipv6 sockets accept ipv4 connects too works for wildcard sockets only. If you want listening on all ip{v4,v6} addresses of an interface you'll need a listening socket for each.

Daniel P. Berrange píše v Pá 15. 06. 2012 v 11:54 +0100: > On Fri, Jun 15, 2012 at 12:44:04PM +0200, Gerd Hoffmann wrote: > > On 06/15/12 12:00, Daniel P. Berrange wrote: > > > On Fri, Jun 15, 2012 at 11:48:28AM +0200, Gerd Hoffmann wrote: > > >> Hi, > > >> > > >>> * new spice-server feature: add option to bind to a selected interface > > >>> regardless of its addresses > > >> > > >> How does that work? > > > > > > I presume the client app would request listen=eth0, and the QEMU > > > would have to call getifaddrs() to determine what IP addresses > > > currently correspond to eth0. > > > > Ah, so there isn't a direct way I'm not aware of, you still bind to a > > specific ip address (or multiple), just specified in a different way ;) > > > > Note that supporting this isn't going to work with a single listening > > socket. Having ipv6 sockets accept ipv4 connects too works for wildcard > > sockets only. If you want listening on all ip{v4,v6} addresses of an > > interface you'll need a listening socket for each. > > Yeah, I'm almost certain you'll need to have multiple listening sockets > for this to work > > Daniel Well, I've checked what my local dnsmasq does and it's doing precisely what you say: tcp 0 0 127.0.0.1:53 0.0.0.0:* LISTEN 29426/dnsmasq tcp 0 0 192.168.122.1:53 0.0.0.0:* LISTEN 29426/dnsmasq tcp 0 0 ::1:53 :::* LISTEN 29426/dnsmasq tcp 0 0 fe80::4c03:d0ff:fec2:aa7:53 :::* LISTEN 29426/dnsmasq In other words, dual-socket won't make dual-stack for selected interface possible without implementing https://bugzilla.redhat.com/show_bug.cgi?id=787256 anyway. :( David