/usr/bin/pg_ctlcluster is in postgresql-common 173ubuntu0.1.
This file is owned by root:root, with mode 0o755.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 | #!/usr/bin/perl -wT
# multiversion/cluster aware pg_ctl wrapper; this also supplies the correct
# configuration parameters to 'start', and makes sure that postgres really
# stops on 'stop'.
#
# (C) 2005-2009 Martin Pitt <mpitt@debian.org>
# (C) 2009 Cyril Bouthors <cyril@bouthors.org>
# (C) 2013-2015 Christoph Berg <myon@debian.org>
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
use strict;
use warnings;
use Getopt::Long;
use POSIX qw/setsid dup2 setlocale LC_ALL :sys_wait_h/;
use PgCommon;
use Fcntl qw(SEEK_SET O_RDWR O_CREAT O_EXCL);
use POSIX qw(lchown);
my ($version, $cluster, $pg_ctl, $force);
my (@postmaster_auxoptions, @pg_ctl_opts_from_cli);
my (%postgresql_conf, %info);
my $mode = 'smart'; # default shutdown mode
my $foreground = 0; # don't daemonize, use postgres instead of pg_ctl
my $stdlog = 0; # when run in foreground, still log to the default logfile
# Check for known broken configurations of this cluster
sub check_valid_config {
my %postgresql_conf = read_cluster_conf_file $version, $cluster,
'postgresql.conf';
my $log_statement_stats = config_bool $postgresql_conf{'log_statement_stats'};
my $log_parser_stats = config_bool $postgresql_conf{'log_parser_stats'};
my $log_planner_stats = config_bool $postgresql_conf{'log_planner_stats'};
my $log_executor_stats = config_bool $postgresql_conf{'log_executor_stats'};
# statement and other statistics are mutually exclusive
if ($log_statement_stats && ($log_parser_stats || $log_planner_stats ||
$log_executor_stats)) {
error ('invalid postgresql.conf: log_statement_stats and the other log_*_stats options are mutually exclusive');
}
}
# If a pid file is already present, delete it if it is stale/invalid, or exit
# with a notice if it belongs to an already running postgres.
sub start_check_pid_file {
my $pidfile = $info{'pgdata'}.'/postmaster.pid';
if (PgCommon::check_pidfile_running $pidfile) {
print "Cluster is already running.\n";
exit 2;
}
# Remove invalid or stale PID file
if (-e $pidfile) {
unlink $pidfile;
print "Removed stale pid file.\n";
}
}
# Check if a pid file is not present or it is invalid. If so, clean up/abort.
sub stop_check_pid_file {
my $pidfile = $info{'pgdata'}.'/postmaster.pid';
my $pid = read_pidfile $pidfile;
return if (defined $pid && PgCommon::check_pidfile_running $pidfile);
if ($info{'running'}) {
error 'pid file is invalid, please manually kill the stale server process.';
}
# Remove invalid or stale PID file
if (-e $pidfile) {
unlink $pidfile;
print "Removed stale pid file.\n";
}
print "Cluster is not running.\n";
exit 2;
}
# check if a cluster reliably connects or fails
# Arguments: <version> <cluster> <port> <socket dir>
sub cluster_port_ready {
my ($v, $c, $p, $sd) = @_;
my $psql = get_program_path 'psql', $v;
error 'cluster_port_ready: could not find psql binary' unless $psql;
my $n = 0;
my $result = 0;
# probe until we get three successful or failed connections in a row
$ENV{'PGPASSWORD'} = 'foo'; # prevent hangs if superuser cannot connect withouth password
my $out;
while ($n < ($result ? 10 : 3)) {
select undef, undef, undef, 0.5;
$out = `$psql -h '$sd' --port $p -l 2>&1 >/dev/null`;
if ($? == $result) {
$n++;
} else {
$n = 0;
}
$result = $?;
}
if ($out =~ 'FATAL:') {
print STDERR "Warning: connection to the database failed, disabling startup checks:\n$out\n";
return cluster_port_running $v, $c, $p;
}
return !$result;
}
sub start {
my $cdir = $info{'configdir'};
check_valid_config;
start_check_pid_file;
# check conflicting port
if (cluster_port_running $version, $cluster, $info{'port'}) {
my $sockdir = get_cluster_socketdir $version, $cluster;
error("Port conflict: another instance is already running on $sockdir with port $info{'port'}");
}
# prepare environment (empty except for content of 'environment', and LANG)
%ENV = read_cluster_conf_file $version, $cluster, 'environment';
# set LANG so non-ascii chars in the server log are not replaced by '?' (affected are non-session contexts)
unless (exists $ENV{LANG}) {
my $lc_messages = PgCommon::get_conf_value $version, $cluster, 'postgresql.conf', 'lc_messages';
$ENV{LANG} = $lc_messages if $lc_messages;
}
# 9.5 configures OOM killer using env vars
if ($version >= 9.5) {
$ENV{PG_OOM_ADJUST_FILE} = "/proc/self/oom_score_adj" unless (exists $ENV{PG_OOM_ADJUST_FILE});
# PG_OOM_ADJUST_VALUE defaults to 0, but can be overridden here
}
my $postmaster_opts = '';
my $usd = $version >= 9.3 ? 'unix_socket_directories' : 'unix_socket_directory';
if (!(PgCommon::get_conf_value $version, $cluster, 'postgresql.conf', $usd)) {
$postmaster_opts .= "-c $usd=\"$info{'socketdir'}\"";
}
$postmaster_opts .= " -c config_file=\"$cdir/postgresql.conf\"";
if (!(PgCommon::get_conf_value $version, $cluster, 'postgresql.conf', 'hba_file')) {
$postmaster_opts .= " -c hba_file=\"$cdir/pg_hba.conf\"";
}
if (!(PgCommon::get_conf_value $version, $cluster, 'postgresql.conf', 'ident_file')) {
$postmaster_opts .= " -c ident_file=\"$cdir/pg_ident.conf\"";
}
if ((-d '/var/run/postgresql') && !defined (PgCommon::get_conf_value $version, $cluster, 'postgresql.conf', 'external_pid_file')) {
# check whether /var/run/postgresql/ is writeable as the cluster owner
my $vrp_writable;
if ($> == 0) {
change_ugid $info{'owneruid'}, $info{'ownergid'};
$vrp_writable = -w '/var/run/postgresql';
$< = $> = 0;
$( = $) = 0;
} else {
$vrp_writable = -w '/var/run/postgresql';
}
if ($vrp_writable) {
$postmaster_opts .= " -c external_pid_file=\"/var/run/postgresql/$version-$cluster.pid\"";
}
}
$postmaster_opts .= ' ' . (join ' ', @postmaster_auxoptions);
($postmaster_opts) = $postmaster_opts =~ /(.*)/; # untaint
if ($foreground) {
if ($stdlog and $info{'logfile'}) {
dup2(POSIX::open($info{logfile}, POSIX::O_WRONLY|POSIX::O_APPEND), 1);
dup2(POSIX::open($info{logfile}, POSIX::O_WRONLY|POSIX::O_APPEND), 2);
}
my $postgres = get_program_path 'postgres', $version;
exec '/bin/sh', '-c', "exec $postgres $postmaster_opts" or
error "Executing $postgres failed: $!"
}
# only supply or default logfile path when none is given explicitly in
# postgresql.conf
my @options = ($pg_ctl, 'start', '-D', $info{'pgdata'});
my $logsize = 0;
if ($info{'logfile'}) {
push @options, ('-l', $info{'logfile'});
# remember current size of the log
$logsize = (stat $info{'logfile'})[7] || 0; # ignore stat errors
}
push @options, @pg_ctl_opts_from_cli if @pg_ctl_opts_from_cli;
my %pg_ctl_opts_from_file = read_cluster_conf_file $version, $cluster, 'pg_ctl.conf';
push @options, split(' ', $pg_ctl_opts_from_file{'pg_ctl_options'})
if defined $pg_ctl_opts_from_file{'pg_ctl_options'} and $pg_ctl_opts_from_file{'pg_ctl_options'} ne '';
push @options, ('-s', '-o', $postmaster_opts);
if (fork) {
wait;
error "could not exec $pg_ctl @options: $!" if $?;
} else {
setsid or error "could not start session: $!";
# without this, you'll get zombies on startup
dup2(POSIX::open('/dev/null', POSIX::O_WRONLY), 1);
dup2(POSIX::open('/dev/null', POSIX::O_WRONLY), 2);
exec $pg_ctl @options or error "could not exec $pg_ctl @options: $!";
}
# wait a bit until the socket exists
my $success = 0;
my $currlogsize = 0;
my $pidfile = $info{'pgdata'}.'/postmaster.pid';
for (my $attempt = 0; $attempt < 60; $attempt++) {
select (undef, undef, undef, 0.5);
$currlogsize = (stat $info{'logfile'})[7] if $info{'logfile'} && -r $info{'logfile'};
if (cluster_port_running $version, $cluster, $info{'port'}) {
$success = 1;
last;
}
# if postgres wrote something, but the process does not exist any
# more, there must be a problem and we can stop immediately
last if ($currlogsize > $logsize && !PgCommon::check_pidfile_running $pidfile);
}
# OK, the server runs, now wait until it stabilized
if ($success) {
$success = cluster_port_ready $version, $cluster, $info{'port'}, $info{'socketdir'};
}
if (!$success) {
if ($currlogsize) {
print STDERR "The PostgreSQL server failed to start. Please check the log output:\n";
open LOG, $info{'logfile'} or
error "Could not open log file " . $info{'logfile'};
seek LOG, $logsize, SEEK_SET;
print STDERR $_ while <LOG>;
} else {
print STDERR "The PostgreSQL server failed to start. Please check the log output.\n";
}
exit 1;
}
return 0;
}
sub stop {
stop_check_pid_file;
my $result = 1;
if ($force) {
if (!fork()) {
close STDOUT;
exec $pg_ctl, '-D', $info{'pgdata'}, '-s', '-w', '-m', 'fast', @pg_ctl_opts_from_cli, 'stop';
} else {
wait;
$result = $? >> 8;
}
# try harder if "fast" mode does not work
if (-f $info{'pgdata'}.'/postmaster.pid') {
print "(does not shutdown gracefully, now stopping immediately)\n";
$result = system $pg_ctl, '-D', $info{'pgdata'}, '-s', '-w', '-m', 'immediate', @pg_ctl_opts_from_cli, 'stop';
}
# if that still not helps, use the big hammer
if (-f $info{'pgdata'}.'/postmaster.pid') {
print "(does not shutdown, killing the process)";
my $pid = read_pidfile ($info{'pgdata'}.'/postmaster.pid');
kill (9, $pid) if $pid;
$result = 0;
}
} else {
if (!fork()) {
close STDOUT;
exec $pg_ctl, '-D', $info{'pgdata'}, '-s', '-w', '-m', $mode, @pg_ctl_opts_from_cli, 'stop';
} else {
wait;
$result = $? >> 8;
}
}
# external_pid_file files are currently not removed by postgres itself
if ($result == 0) {
unlink "/var/run/postgresql/$version-$cluster.pid";
}
return $result;
}
sub restart {
my $result;
if ($info{'running'}) {
$result = stop;
return $result if $result;
}
return start;
}
sub reload {
exec $pg_ctl, '-D', $info{'pgdata'}, '-s', @pg_ctl_opts_from_cli, 'reload';
}
sub status {
exec $pg_ctl, '-D', $info{'pgdata'}, 'status';
}
sub promote {
exec $pg_ctl, '-D', $info{'pgdata'}, '-s', @pg_ctl_opts_from_cli, 'promote';
}
#
# main
#
my ($skip_systemctl_redirect, $bindir);
exit 1 unless GetOptions ('o|options=s' => \@postmaster_auxoptions,
'f|force' => \$force,
'm|mode=s' => \$mode,
'foreground' => \$foreground,
'skip-systemctl-redirect' => \$skip_systemctl_redirect,
'stdlog' => \$stdlog,
'bindir=s' => \$bindir,
);
if ($mode =~ /^(s(mart)?|f(ast)?|i(mmediate)?)$/) {
$mode = $1; # untaint
} else {
print "Invalid -m mode, valid are: smart fast immediate\n";
exit 1;
}
($bindir) = $bindir =~ /^(\/.*)$/ if $bindir; # untaint
# accept both "version cluster action" and "version[-/]cluster action"
if (@ARGV < 2 or (@ARGV < 3 and $ARGV[0] !~ m![-/]!)) {
print "Usage: $0 <version> <cluster> <action>\n";
exit 1;
}
$version = shift @ARGV;
if ($version =~ m!^(\d+\.\d)[-/](.+)!) {
($version, $cluster) = ($1, $2);
} else {
$cluster = shift @ARGV;
}
my $action = shift @ARGV;
@pg_ctl_opts_from_cli=();
foreach my $argv (@ARGV) {
push @pg_ctl_opts_from_cli, $argv =~ /(.*)/; # untaint
}
($version) = $version =~ /^(\d+\.\d+)$/; # untaint
($cluster) = $cluster =~ /^([^'"\s]+)$/; # untaint
error 'specified cluster does not exist' unless $version && $cluster && cluster_exists $version, $cluster;
%info = cluster_info ($version, $cluster);
unless ($action eq 'stop') {
# check if cluster is disabled in start.conf
error 'Cluster is disabled' if $info{'start'} eq 'disabled';
}
# untaint environment
$ENV{'PATH'} = '/sbin:/bin:/usr/sbin:/usr/bin';
delete @ENV{'IFS', 'CDPATH', 'ENV', 'BASH_ENV'};
unless (-d $info{'pgdata'} && defined $info{'owneruid'}) {
error $info{'pgdata'} . ' is not accessible or does not exist';
}
# check that owner uid/gid is valid
unless (getpwuid $info{'owneruid'}) {
error 'The cluster is owned by user id '.$info{'owneruid'}.' which does not exist'
}
unless (getgrgid $info{'ownergid'}) {
error 'The cluster is owned by group id '.$info{'ownergid'}.' which does not exist'
}
# owneruid and configuid need to match, unless configuid is root
if (($< == 0 or $> == 0) and $info{'configuid'} != 0 and
$info{'configuid'} != $info{'owneruid'}) {
my $configowner = (getpwuid $info{'configuid'})[0] || "(unknown)";
my $dataowner = (getpwuid $info{'owneruid'})[0];
error "Config owner ($configowner:$info{configuid}) and data owner ($dataowner:$info{owneruid}) do not match, and config owner is not root";
}
# redirect the request through systemd
if (not $skip_systemctl_redirect and getppid() != 1 and # not run from init
-d '/run/systemd/system' and not exists $ENV{_SYSTEMCTL_SKIP_REDIRECT} and # systemd is running
not exists $ENV{'PG_CLUSTER_CONF_ROOT'} and # not handling user clusters
$action =~ /^(start|stop|restart)$/ # redirect only these actions
) {
$action = $1; # untaint
system 'systemctl', 'is-active', '-q', "postgresql\@$version-$cluster";
my $unit_active = $? == 0;
# if extra options are given, proceed with pg_ctlcluster with a warning
if (@postmaster_auxoptions != 0 or @pg_ctl_opts_from_cli != 0 or $bindir) { # extra options given
print "Notice: extra pg_ctl/postgres options given, bypassing systemctl for $action operation\n" if (-t 1);
# if we are root, redirect to systemctl unless stopping a cluster running outside systemd
} elsif ($> == 0) {
if ($action eq 'stop' and not $unit_active) {
# proceed with pg_ctlcluster
} else {
print "Redirecting $action request to systemctl\n" if (-t 1);
system 'systemctl', $action, "postgresql\@$version-$cluster";
exit $? >> 8;
# program end
}
# as non-root, raise an error on restarting a cluster running from systemd
} elsif ($action eq 'restart' and $unit_active) {
error "cluster is running from systemd, can only restart it as root. Try instead:\n sudo systemctl $action postgresql\@$version-$cluster";
# program end
# otherwise just raise a warning on start and restart as non-root
} elsif (-t 1) {
if ($action =~ /start/) {
print "Warning: the cluster will not be running as a systemd service. Consider using systemctl:\n";
} elsif ($unit_active) { # on stop, warn when running from systemd
print "Warning: stopping the cluster using pg_ctlcluster will mark the systemd unit as failed. Consider using systemctl:\n";
}
print " sudo systemctl $action postgresql\@$version-$cluster\n";
}
}
# recreate missing standard log dir
if ($> == 0 && ! -e '/var/log/postgresql' &&
$info{'logfile'} =~ m!^/var/log/postgresql!) {
system 'install', '-d', '-m', '1775', '-o', 'root', '-g', 'postgres', '/var/log/postgresql';
}
# recreate missing log file
if ($action ne 'stop' && $info{'logfile'} && ! -e $info{'logfile'}) {
if ($> == 0) { # drop privileges; this is important if logfile
# was determined via an /etc/postgresql/.../log symlink
change_ugid $info{'owneruid'}, $info{'ownergid'};
}
sysopen (L, $info{'logfile'}, O_RDWR|O_CREAT|O_EXCL) or
error 'Could not create log file ' . $info{'logfile'};
close L;
chmod 0640, $info{'logfile'};
$< = $> = 0; # will silently fail if we were not root before, that's intended
$( = $) = 0;
if ($info{'owneruid'} < 500) {
my $g = (getgrnam 'adm')[2];
lchown $info{'owneruid'}, $g, $info{'logfile'} if (defined $g);
}
}
# recreate /var/run/postgresql
if ($action ne 'stop' && ! -d $info{'socketdir'}) {
system 'install', '-d', '-m', 2775,
'-o', $info{'owneruid'}, '-g', $info{'ownergid'}, $info{'socketdir'};
}
# recreate stats_temp_directory
if ($action ne 'stop' && $info{'statstempdir'} && ! -d $info{'statstempdir'}) {
system 'install', '-d', '-m', 750,
'-o', $info{'owneruid'}, '-g', $info{'ownergid'}, $info{'statstempdir'};
}
if ($> == 0) {
# have postgres start with increased OOM killer protection; 9.0 and
# later has builtin support for resetting the adjustment of child processes
if ($action eq 'start' and $version >= '9.0' and not $PgCommon::rpm) {
if (-w '/proc/self/oom_score_adj') {
open F, '>/proc/self/oom_score_adj';
print F "-900\n";
close F;
}
}
chdir ('/var/lib/postgresql');
change_ugid $info{'owneruid'}, $info{'ownergid'};
}
# we are running as the cluster owner now
if( $> != $info{'owneruid'} ) {
error 'You must run this program as the cluster owner ('.
(getpwuid $info{'owneruid'})[0].') or root';
}
$pg_ctl = $bindir ? "$bindir/pg_ctl" : get_program_path ('pg_ctl', $version);
if ($action =~ /^(start|stop|restart|reload|status|promote)$/) {
no strict 'refs';
exit &$action;
} else {
error 'Invalid action (must be one of: start stop restart reload status promote)';
}
__END__
=head1 NAME
pg_ctlcluster - start/stop/restart/reload a PostgreSQL cluster
=head1 SYNOPSIS
B<pg_ctlcluster> [I<options>] I<cluster-version> I<cluster-name> I<action> B<--> [I<pg_ctl options>]
where I<action> = B<start>|B<stop>|B<restart>|B<reload>|B<status>|B<promote>
=head1 DESCRIPTION
This program controls the B<postgres> server for a particular cluster. It
essentially wraps the L<pg_ctl(1)> command. It determines the cluster version
and data path and calls the right version of B<pg_ctl> with appropriate
configuration parameters and paths.
You have to start this program as the user who owns the database cluster or as
root.
To ease integration with B<systemd> operation, the alternative syntax
"B<pg_ctlcluster> I<version>B<->I<cluster> I<action>" is also supported.
=head1 ACTIONS
=over 4
=item B<start>
A log file for this specific cluster is created if it does not exist yet (by
default,
C</var/log/postgresql/postgresql->I<cluster-version>C<->I<cluster-name>C<.log>),
and a PostreSQL server process (L<postgres(1)>) is started on it. Exits with
0 on success, with 2 if the server is already running, and with 1 on other
failure conditions.
=item B<stop>
Stops the L<postgres(1)> server of the given cluster. By default, "smart"
shutdown mode is used, which waits until all clients disconnected.
=item B<restart>
Stops the server if it is running and starts it (again).
=item B<reload>
Causes the configuration files to be re-read without a full shutdown of the
server.
=item B<status>
Checks whether a server is running. If it is, the PID and the command line
options that were used to invoke it are displayed.
=item B<promote>
Commands a running standby server to exit recovery and begin read-write
operations.
=back
=head1 OPTIONS
=over 4
=item B<-f>|B<--force>
For B<stop> and B<restart>, the "fast" mode is used which rolls back all active
transactions, disconnects clients immediately and thus shuts down cleanly. If
that does not work, shutdown is attempted again in "immediate" mode, which can
leave the cluster in an inconsistent state and thus will lead to a recovery run
at the next start. If this still does not help, the B<postgres> process is
killed. Exits with 0 on success, with 2 if the server is not running, and with
1 on other failure conditions. This mode should only be used when the machine
is about to be shut down.
=item B<-m>|B<--mode> [B<smart>|B<fast>|B<immediate>]
Shutdown mode to use for B<stop> and B<restart> actions, default is B<smart>.
See pg_ctl(1) for documentation.
=item B<--foreground>
Start B<postgres> in foreground, without daemonizing via B<pg_ctl>.
=item B<--stdlog>
When B<--foreground> is in use, redirect stderr to the standard logfile in
C</var/log/postgresql/>. (Default when not run in foreground.)
=item B<--bindir> I<directory>
Path to B<pg_ctl>. (Default is C</usr/lib/postgresql/>I<version>C</bin>.)
=item B<-o>|B<--options> I<option>
Pass given I<option> as command line option to the C<postgres> process. It is
possible to specify B<-o> multiple times. See L<postgres(1)> for a
description of valid options.
=item I<pg_ctl options>
Pass given I<pg_ctl options> as command line options to B<pg_ctl>. See L<pg_ctl(1)>
for a description of valid options.
=back
=head1 FILES
=over 4
=item C</etc/postgresql/>I<cluster-version>C</>I<cluster-name>C</pg_ctl.conf>
This configuration file contains cluster specific options to be passed to
L<pg_ctl(1)>.
=item C</etc/postgresql/>I<cluster-version>C</>I<cluster-name>C</start.conf>
This configuration file controls the start/stop behavior of the cluster. See
section "STARTUP CONTROL" in L<pg_createcluster(8)> for details.
=back
=head1 BUGS
Changing the port number on startup using B<-o -p> will not work as it breaks
the checks for running clusters.
=head1 SEE ALSO
L<pg_createcluster(8)>, L<pg_ctl(1)>, L<pg_wrapper(1)>, L<pg_lsclusters(1)>,
L<postgres(1)>
=head1 AUTHOR
Martin Pitt L<E<lt>mpitt@debian.orgE<gt>>
|