amavisd-new-2.7.0 is the current stable version upstream as of july 2011. Please stabilize it and please don't let it get years old like the 2.6.x series in portage :( We are starting internal testing on gentoo/hardened/amd64 and I'll report back any issue.
Arches, please test and mark stable =mail-filter/amavisd-new-2.7.0. Thank you.
We switched to this on our main MX a while ago for the before-queue improvements. You have my vote for x86.
amd64: ok
+ 08 Sep 2011; Tony Vroon <chainsaw@gentoo.org> amavisd-new-2.7.0.ebuild: + Marked stable on AMD64 based on arch testing by Ian "idella4" Delaney in bug + #381977 filed by Luca Lesinigo.
It's broken with 2.7.0 as it doesn't start because of this.... Subroutine post_bind_hook redefined at /usr/sbin/amavisd line 10841. Looking at /usr/sbin/amavisd we have two definitions.... ### Net::Server hook ### Occurs in the parent (master) process after binding to sockets, ### but before chrooting and dropping privileges # sub post_bind_hook { umask(0027); # restore our preferred umask set_sockets_access() if defined $warm_restart && !$warm_restart; } ### Net::Server hook ### This hook occurs just after the bind process and just before any ### chrooting, change of user, or change of group occurs. At this point ### the process will still be running as the user who started the server. sub post_bind_hook { my ($self) = @_; if (c('protocol') eq 'COURIER') { # Allow courier to write to the socket chmod(0660, $unix_socketname); } if ($self->{courierfilter_pipe}) { # Watch for courierfilter telling us to shut down $self->{server}->{select}->add($self->{courierfilter_pipe}); } }
Is that with USE="courier"? Looking at mine without that just has the first definition.
Yes, that's correct.
Stable for HPPA.
x86 stable
reverted back to ~ppc/~ppc64, done
Stable on alpha.
sparc stable, closing