www: add configurable limiters

Currently only for git-http-backend use, this allows limiting
the number of spawned processes per-inbox or by group, if there
are multiple large inboxes amidst a sea of small ones.

For example, a "big" repo limiter could be used for big inboxes:
which would be shared between multiple repos:

[limiter "big"]
	max = 4
[publicinbox "git"]
	address = git@vger.kernel.org
	mainrepo = /path/to/git.git
	; shared limiter with giant:
	httpbackendmax = big
[publicinbox "giant"]
	address = giant@project.org
	mainrepo = /path/to/giant.git
	; shared limiter with git:
	httpbackendmax = big

; This is a tiny inbox, use the default limiter with 32 slots:
[publicinbox "meta"]
	address = meta@public-inbox.org
	mainrepo = /path/to/meta.git

diff --git a/MANIFEST b/MANIFEST
index ceb1a9dadf9fc74ed0234eb647397e56be567d7e..75bb43e35bbe1210dca1698bed00eddd95d600c4 100644 (file)
--- a/MANIFEST
+++ b/MANIFEST
@@ -106,6 +106,7 @@ t/cgi.t
 t/check-www-inbox.perl
 t/common.perl
 t/config.t
+t/config_limiter.t
 t/emergency.t
 t/fail-bin/spamc
 t/feed.t

diff --git a/lib/PublicInbox/Config.pm b/lib/PublicInbox/Config.pm
index d34d11ad5cee2a3ef1d4334e8dc94b044b4e5e85..d7eaa3e8f1c5100279deaefc7a0a8f1973023c5d 100644 (file)
--- a/lib/PublicInbox/Config.pm
+++ b/lib/PublicInbox/Config.pm
@@ -20,6 +20,7 @@ sub new {
        $self->{-by_addr} ||= {};
        $self->{-by_name} ||= {};
        $self->{-by_newsgroup} ||= {};
+       $self->{-limiters} ||= {};
        $self;
 }
 
@@ -85,6 +86,15 @@ sub lookup_newsgroup {
        undef;
 }
 
+sub limiter {
+       my ($self, $name) = @_;
+       $self->{-limiters}->{$name} ||= do {
+               require PublicInbox::Qspawn;
+               my $key = "limiter.$name.max";
+               PublicInbox::Qspawn::Limiter->new($self->{$key});
+       };
+}
+
 sub get {
        my ($self, $inbox, $key) = @_;
 
@@ -131,7 +141,7 @@ sub _fill {
        my $rv = {};
 
        foreach my $k (qw(mainrepo address filter url newsgroup
-                       watch watchheader)) {
+                       watch watchheader httpbackendmax)) {
                my $v = $self->{"$pfx.$k"};
                $rv->{$k} = $v if defined $v;
        }
@@ -139,6 +149,7 @@ sub _fill {
        my $name = $pfx;
        $name =~ s/\Apublicinbox\.//;
        $rv->{name} = $name;
+       $rv->{-pi_config} = $self;
        $rv = PublicInbox::Inbox->new($rv);
        my $v = $rv->{address};
        if (ref($v) eq 'ARRAY') {

diff --git a/lib/PublicInbox/GitHTTPBackend.pm b/lib/PublicInbox/GitHTTPBackend.pm
index ed8fdf004c6ced412c1238ef8d10250e2402d6c0..d4914795f3ebfbacc206466721a4e27d0a9a38f7 100644 (file)
--- a/lib/PublicInbox/GitHTTPBackend.pm
+++ b/lib/PublicInbox/GitHTTPBackend.pm
@@ -179,7 +179,6 @@ sub prepare_range {
 # returns undef if 403 so it falls back to dumb HTTP
 sub serve_smart {
        my ($env, $git, $path) = @_;
-       my $limiter = $default_limiter;
        my $in = $env->{'psgi.input'};
        my $fd = eval { fileno($in) };
        unless (defined $fd && $fd >= 0) {
@@ -197,7 +196,14 @@ sub serve_smart {
                my $val = $env->{$name};
                $env{$name} = $val if defined $val;
        }
-       my $git_dir = ref $git ? $git->{git_dir} : $git;
+       my ($git_dir, $limiter);
+       if (ref $git) {
+               $limiter = $git->{-httpbackend_limiter} || $default_limiter;
+               $git_dir = $git->{git_dir};
+       } else {
+               $limiter = $default_limiter;
+               $git_dir = $git;
+       }
        $env{GIT_HTTP_EXPORT_ALL} = '1';
        $env{PATH_TRANSLATED} = "$git_dir/$path";
        my %rdr = ( 0 => fileno($in) );

diff --git a/lib/PublicInbox/Inbox.pm b/lib/PublicInbox/Inbox.pm
index dc9980b77c62b0db75b09b58897c40a17f1e0e16..23b77213b7a9d1b3adf4b8598ddca22035dea4a3 100644 (file)
--- a/lib/PublicInbox/Inbox.pm
+++ b/lib/PublicInbox/Inbox.pm
@@ -34,6 +34,7 @@ sub new {
        my $v = $opts->{address} ||= 'public-inbox@example.com';
        my $p = $opts->{-primary_address} = ref($v) eq 'ARRAY' ? $v->[0] : $v;
        $opts->{domain} = ($p =~ /\@(\S+)\z/) ? $1 : 'localhost';
+       weaken($opts->{-pi_config});
        bless $opts, $class;
 }
 
@@ -44,11 +45,33 @@ sub _weaken_fields {
        }
 }
 
+sub _set_limiter ($$$) {
+       my ($self, $git, $pfx) = @_;
+       my $lkey = "-${pfx}_limiter";
+       $git->{$lkey} = $self->{$lkey} ||= eval {
+               my $mkey = $pfx.'max';
+               my $val = $self->{$mkey} or return;
+               my $lim;
+               if ($val =~ /\A\d+\z/) {
+                       require PublicInbox::Qspawn;
+                       $lim = PublicInbox::Qspawn::Limiter->new($val);
+               } elsif ($val =~ /\A[a-z][a-z0-9]*\z/) {
+                       $lim = $self->{-pi_config}->limiter($val);
+                       warn "$mkey limiter=$val not found\n" if !$lim;
+               } else {
+                       warn "$mkey limiter=$val not understood\n";
+               }
+               $lim;
+       }
+}
+
 sub git {
        my ($self) = @_;
        $self->{git} ||= eval {
                _weaken_later($self);
-               PublicInbox::Git->new($self->{mainrepo});
+               my $g = PublicInbox::Git->new($self->{mainrepo});
+               _set_limiter($self, $g, 'httpbackend');
+               $g;
        };
 }
 

diff --git a/lib/PublicInbox/Qspawn.pm b/lib/PublicInbox/Qspawn.pm
index cc9c340da26530993267f8cd00dc30da0d83714f..697c55a1b4d5ec8ad579a03a0b8fdea6e45136a2 100644 (file)
--- a/lib/PublicInbox/Qspawn.pm
+++ b/lib/PublicInbox/Qspawn.pm
@@ -47,7 +47,7 @@ sub start {
        my ($self, $limiter, $cb) = @_;
        $self->{limiter} = $limiter;
 
-       if ($limiter->{running} < $limiter->{limit}) {
+       if ($limiter->{running} < $limiter->{max}) {
                _do_spawn($self, $cb);
        } else {
                push @{$limiter->{run_queue}}, [ $self, $cb ];
@@ -59,9 +59,10 @@ use strict;
 use warnings;
 
 sub new {
-       my ($class, $limit) = @_;
+       my ($class, $max) = @_;
        bless {
-               limit => $limit || 1,
+               # 32 is same as the git-daemon connection limit
+               max => $max || 32,
                running => 0,
                run_queue => [],
        }, $class;

diff --git a/t/config.t b/t/config.t
index dc448cdffff6b7f92574a4a7903323ccd9e769b3..77e8f4ac18e5bd7b8b153a18326e12a5581690e6 100644 (file)
--- a/t/config.t
+++ b/t/config.t
@@ -30,6 +30,7 @@ my $tmpdir = tempdir('pi-config-XXXXXX', TMPDIR => 1, CLEANUP => 1);
                'url' => 'http://example.com/meta',
                -primary_address => 'meta@public-inbox.org',
                'name' => 'meta',
+               -pi_config => $cfg,
        }, "lookup matches expected output");
 
        is($cfg->lookup('blah@example.com'), undef,
@@ -45,6 +46,7 @@ my $tmpdir = tempdir('pi-config-XXXXXX', TMPDIR => 1, CLEANUP => 1);
                'domain' => 'public-inbox.org',
                'name' => 'test',
                'url' => 'http://example.com/test',
+               -pi_config => $cfg,
        }, "lookup matches expected output for test");
 }
 

diff --git a/t/config_limiter.t b/t/config_limiter.t
new file mode 100644 (file)
index 0000000..bfea151
--- /dev/null
+++ b/t/config_limiter.t
@@ -0,0 +1,50 @@
+# Copyright (C) 2016 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+use strict;
+use warnings;
+use Test::More;
+use PublicInbox::Config;
+my $cfgpfx = "publicinbox.test";
+{
+       my $config = PublicInbox::Config->new({
+               "$cfgpfx.address" => 'test@example.com',
+               "$cfgpfx.mainrepo" => '/path/to/non/existent',
+               "$cfgpfx.httpbackendmax" => 12,
+       });
+       my $ibx = $config->lookup_name('test');
+       my $git = $ibx->git;
+       my $old = "$git";
+       my $lim = $git->{-httpbackend_limiter};
+       ok($lim, 'Limiter exists');
+       is($lim->{max}, 12, 'limiter has expected slots');
+       $git = undef;
+       $ibx->{git} = undef;
+       $git = $ibx->git;
+       isnt($old, "$git", 'got new Git object');
+       is("$git->{-httpbackend_limiter}", "$lim", 'same limiter');
+}
+
+{
+       my $config = PublicInbox::Config->new({
+               'limiter.named.max' => 3,
+               "$cfgpfx.address" => 'test@example.com',
+               "$cfgpfx.mainrepo" => '/path/to/non/existent',
+               "$cfgpfx.httpbackendmax" => 'named',
+       });
+       my $ibx = $config->lookup_name('test');
+       my $git = $ibx->git;
+       ok($git, 'got git object');
+       my $old = "$git";
+       my $lim = $git->{-httpbackend_limiter};
+       ok($lim, 'Limiter exists');
+       is($lim->{max}, 3, 'limiter has expected slots');
+       $git = undef;
+       $ibx->{git} = undef;
+       PublicInbox::Inbox::weaken_task;
+       $git = $ibx->git;
+       isnt($old, "$git", 'got new Git object');
+       is("$git->{-httpbackend_limiter}", "$lim", 'same limiter');
+       is($lim->{max}, 3, 'limiter has expected slots');
+}
+
+done_testing;