user/dev discussion of public-inbox itself
 help / color / Atom feed
From: Eric Wong <e@yhbt.net>
To: meta@public-inbox.org
Subject: [PATCH 3/3] xapcmd: increase scope of lock
Date: Sun, 26 Jan 2020 01:17:44 +0000
Message-ID: <20200126011744.6278-4-e@yhbt.net> (raw)
In-Reply-To: <20200126011744.6278-1-e@yhbt.net>

The old lock scope was only sufficient for protecting against
concurrent modifications from the common -mda, -watch, or -learn
writers.

It was not sufficient for protecting against parallel -compact
or -xcpdb invocations from eager admins.  Most of the time this
only leads to confusing and misleading warning messages, but
parallel xcpdb --reshard could lead to errors.
---
 lib/PublicInbox/Xapcmd.pm | 59 +++++++++++++++++++++------------------
 1 file changed, 32 insertions(+), 27 deletions(-)

diff --git a/lib/PublicInbox/Xapcmd.pm b/lib/PublicInbox/Xapcmd.pm
index 19c6ff07..2864b0d4 100644
--- a/lib/PublicInbox/Xapcmd.pm
+++ b/lib/PublicInbox/Xapcmd.pm
@@ -122,7 +122,8 @@ sub same_fs_or_die ($$) {
 }
 
 sub process_queue {
-	my ($queue, $cb, $max, $opt) = @_;
+	my ($queue, $cb, $opt) = @_;
+	my $max = $opt->{jobs} || scalar(@$queue);
 	if ($max <= 1) {
 		while (defined(my $args = shift @$queue)) {
 			$cb->($args, $opt);
@@ -152,36 +153,18 @@ sub setup_signals () {
 	$SIG{HUP} = $SIG{PIPE} = $SIG{TERM} = sub { exit(1) };
 }
 
-sub run {
-	my ($ibx, $task, $opt) = @_; # task = 'cpdb' or 'compact'
-	my $cb = \&${\"PublicInbox::Xapcmd::$task"};
-	PublicInbox::Admin::progress_prepare($opt ||= {});
-	my $dir = $ibx->{inboxdir} or die "no inboxdir in inbox\n";
-	runnable_or_die($XAPIAN_COMPACT) if $opt->{compact};
-	my $reindex; # v1:{ from => $x40 }, v2:{ from => [ $x40, $x40, .. ] } }
-	my $from; # per-epoch ranges
-
-	if (!$opt->{-coarse_lock}) {
-		$reindex = $opt->{reindex} = {};
-		$from = $reindex->{from} = [];
-		require PublicInbox::SearchIdx;
-		PublicInbox::SearchIdx::load_xapian_writable();
-	}
+sub prepare_run {
+	my ($ibx, $opt) = @_;
+	my $tmp = {}; # old shard dir => File::Temp->newdir object or undef
+	my @queue; # ([old//src,newdir]) - list of args for cpdb() or compact()
 
-	$ibx->umask_prepare;
 	my $old = $ibx->search->xdir(1);
 	-d $old or die "$old does not exist\n";
-
-	my $tmp = {};
-	my @q;
 	my $reshard = $opt->{reshard};
 	if (defined $reshard && $reshard <= 0) {
 		die "--reshard must be a positive number\n";
 	}
 
-	local %SIG = %SIG;
-	setup_signals();
-
 	# we want temporary directories to be as deep as possible,
 	# so v2 shards can keep "xap$SCHEMA_VERSION" on a separate FS.
 	if ($ibx->version == 1) {
@@ -194,7 +177,7 @@ sub run {
 		my $v = PublicInbox::Search::SCHEMA_VERSION();
 		my $wip = File::Temp->newdir("xapian$v-XXXXXXXX", DIR => $dir);
 		$tmp->{$old} = $wip;
-		push @q, [ $old, $wip ];
+		push @queue, [ $old, $wip ];
 	} else {
 		opendir my $dh, $old or die "Failed to opendir $old: $!\n";
 		my @old_shards;
@@ -223,7 +206,7 @@ sub run {
 			my $wip = File::Temp->newdir($tmpl, DIR => $old);
 			same_fs_or_die($old, $wip->dirname);
 			my $cur = "$old/$dn";
-			push @q, [ $src // $cur , $wip ];
+			push @queue, [ $src // $cur , $wip ];
 			$tmp->{$cur} = $wip;
 		}
 		# mark old shards to be unlinked
@@ -231,10 +214,32 @@ sub run {
 			$tmp->{$_} ||= undef for @$src;
 		}
 	}
-	my $max = $opt->{jobs} || scalar(@q);
+	($tmp, \@queue);
+}
+
+sub run {
+	my ($ibx, $task, $opt) = @_; # task = 'cpdb' or 'compact'
+	my $cb = \&${\"PublicInbox::Xapcmd::$task"};
+	PublicInbox::Admin::progress_prepare($opt ||= {});
+	defined(my $dir = $ibx->{inboxdir}) or die "no inboxdir defined\n";
+	-d $dir or die "inboxdir=$dir does not exist\n";
+	runnable_or_die($XAPIAN_COMPACT) if $opt->{compact};
+	my $reindex; # v1:{ from => $x40 }, v2:{ from => [ $x40, $x40, .. ] } }
+
+	if (!$opt->{-coarse_lock}) {
+		$reindex = $opt->{reindex} = {};
+		$reindex->{from} = []; # per-epoch ranges
+		require PublicInbox::SearchIdx;
+		PublicInbox::SearchIdx::load_xapian_writable();
+	}
+
+	local %SIG = %SIG;
+	setup_signals();
+	$ibx->umask_prepare;
 	$ibx->with_umask(sub {
 		my $im = $ibx->importer(0);
 		$im->lock_acquire;
+		my ($tmp, $queue) = prepare_run($ibx, $opt);
 
 		# fine-grained locking if we prepare for reindex
 		if (!$opt->{-coarse_lock}) {
@@ -243,7 +248,7 @@ sub run {
 		}
 
 		$ibx->cleanup;
-		process_queue(\@q, $cb, $max, $opt);
+		process_queue($queue, $cb, $opt);
 		$im->lock_acquire if !$opt->{-coarse_lock};
 		commit_changes($ibx, $im, $tmp, $opt);
 	});

      parent reply index

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-26  1:17 [PATCH 0/3] writing/admin code cleanups lock fix Eric Wong
2020-01-26  1:17 ` [PATCH 1/3] inbox: add ->version method Eric Wong
2020-01-26  1:17 ` [PATCH 2/3] search: {version} => {ibx_ver} Eric Wong
2020-01-26  1:17 ` Eric Wong [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://public-inbox.org/README

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200126011744.6278-4-e@yhbt.net \
    --to=e@yhbt.net \
    --cc=meta@public-inbox.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

user/dev discussion of public-inbox itself

Archives are clonable:
	git clone --mirror https://public-inbox.org/meta
	git clone --mirror http://czquwvybam4bgbro.onion/meta
	git clone --mirror http://hjrcffqmbrq6wope.onion/meta
	git clone --mirror http://ou63pmih66umazou.onion/meta

Example config snippet for mirrors

Newsgroups are available over NNTP:
	nntp://news.public-inbox.org/inbox.comp.mail.public-inbox.meta
	nntp://ou63pmih66umazou.onion/inbox.comp.mail.public-inbox.meta
	nntp://czquwvybam4bgbro.onion/inbox.comp.mail.public-inbox.meta
	nntp://hjrcffqmbrq6wope.onion/inbox.comp.mail.public-inbox.meta
	nntp://news.gmane.io/gmane.mail.public-inbox.general

 note: .onion URLs require Tor: https://www.torproject.org/

AGPL code for this site: git clone https://public-inbox.org/public-inbox.git