1 # Copyright (C) 2018-2020 all contributors <meta@public-inbox.org>
2 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
4 # Internal interface for a single Xapian shard in V2 inboxes.
5 # See L<public-inbox-v2-format(5)> for more info on how we shard Xapian
6 package PublicInbox::SearchIdxShard;
9 use parent qw(PublicInbox::SearchIdx);
10 use IO::Handle (); # autoflush
14 my ($class, $v2w, $shard) = @_;
15 my $ibx = $v2w->{ibx};
16 my $self = $class->SUPER::new($ibx, 1, $shard);
17 # create the DB before forking:
19 $self->set_metadata_once;
21 $self->spawn_worker($v2w, $shard) if $v2w->{parallel};
26 my ($self, $v2w, $shard) = @_;
28 pipe($r, $w) or die "pipe failed: $!\n";
31 defined $pid or die "fork failed: $!\n";
33 my $bnote = $v2w->atfork_child;
34 close $w or die "failed to close: $!";
36 # F_SETPIPE_SZ = 1031 on Linux; increasing the pipe size here
37 # speeds V2Writable batch imports across 8 cores by nearly 20%
38 fcntl($r, 1031, 1048576) if $^O eq 'linux';
40 eval { shard_worker_loop($self, $v2w, $r, $shard, $bnote) };
41 die "worker $shard died: $@\n" if $@;
42 die "unexpected MM $self->{mm}" if $self->{mm};
47 close $r or die "failed to close: $!";
50 # this reads all the writes to $self->{w} from the parent process
51 sub shard_worker_loop ($$$$$) {
52 my ($self, $v2w, $r, $shard, $bnote) = @_;
53 $0 = "pi-v2-shard[$shard]";
54 $self->begin_txn_lazy;
55 while (my $line = readline($r)) {
56 $v2w->{current_info} = "[$shard] $line";
57 if ($line eq "commit\n") {
58 $self->commit_txn_lazy;
59 } elsif ($line eq "close\n") {
61 } elsif ($line eq "barrier\n") {
62 $self->commit_txn_lazy;
63 # no need to lock < 512 bytes is atomic under POSIX
64 print $bnote "barrier $shard\n" or
65 die "write failed for barrier $!\n";
66 } elsif ($line =~ /\AD ([a-f0-9]{40,}) ([0-9]+)\n\z/s) {
67 $self->remove_by_oid($1, $2 + 0);
70 # n.b. $mid may contain spaces(!)
71 my ($to_read, $bytes, $num, $blob, $ds, $ts, $tid, $mid)
72 = split(/ /, $line, 8);
73 $self->begin_txn_lazy;
74 my $n = read($r, my $msg, $to_read) or die "read: $!\n";
75 $n == $to_read or die "short read: $n != $to_read\n";
76 my $mime = PublicInbox::Eml->new(\$msg);
85 }, 'PublicInbox::Smsg';
86 $self->add_message($mime, $smsg);
93 my ($self, $msgref, $eml, $smsg) = @_;
94 if (my $w = $self->{w}) {
95 # mid must be last, it can contain spaces (but not LF)
96 print $w join(' ', @$smsg{qw(raw_bytes bytes
97 num blob ds ts tid mid)}),
98 "\n", $$msgref or die "failed to write shard $!\n";
102 } else { # --xapian-only + --sequential-shard:
103 $eml = PublicInbox::Eml->new($msgref);
105 $self->begin_txn_lazy;
106 $self->add_message($eml, $smsg);
111 close $_[0]->{w} or die "failed to close write pipe: $!\n";
116 if (my $w = $self->{w}) {
117 print $w "barrier\n" or die "failed to print: $!";
119 $self->commit_txn_lazy;
125 if (my $w = $self->{w}) {
126 print $w "commit\n" or die "failed to write commit: $!";
128 $self->commit_txn_lazy;
134 if (my $w = delete $self->{w}) {
135 my $pid = delete $self->{pid} or die "no process to wait on\n";
136 print $w "close\n" or die "failed to write to pid:$pid: $!\n";
137 close $w or die "failed to close pipe for pid:$pid: $!\n";
138 waitpid($pid, 0) == $pid or die "remote process did not finish";
139 $? == 0 or die ref($self)." pid:$pid exited with: $?";
141 die "transaction in progress $self\n" if $self->{txn};
142 $self->idx_release if $self->{xdb};
147 my ($self, $oid, $num) = @_;
148 if (my $w = $self->{w}) { # triggers remove_by_oid in a shard child
149 print $w "D $oid $num\n" or die "failed to write remove $!";
150 } else { # same process
151 $self->remove_by_oid($oid, $num);