| Current File : //usr/local/nagios/libexec/check_postgres |
#!/usr/bin/perl -- -*-cperl-*-
## Perform many different checks against Postgres databases.
## Designed primarily as a Nagios script.
## Run with --help for a summary.
##
## Greg Sabino Mullane <greg@endpoint.com>
## End Point Corporation http://www.endpoint.com/
## BSD licensed, see complete license at bottom of this script
## The latest version can be found at:
## http://www.bucardo.org/check_postgres/
##
## See the HISTORY section for other contributors
package check_postgres;
use v5.6.0;
use strict;
use warnings;
use Getopt::Long qw/GetOptions/;
Getopt::Long::Configure('no_ignore_case');
use File::Basename qw/basename/;
use File::Temp qw/tempfile tempdir/;
File::Temp->safe_level( File::Temp::MEDIUM ); ## no critic
use Cwd;
use Data::Dumper qw/Dumper/;
$Data::Dumper::Varname = 'POSTGRES';
$Data::Dumper::Indent = 2;
$Data::Dumper::Useqq = 1;
our $VERSION = '1.5.2';
use vars qw/ %opt $PSQL $res $COM $SQL $db /;
## If psql is not in your path, it is recommended that hardcode it here,
## as an alternative to the --PSQL option
$PSQL = '';
## If this is true, $opt{PSQL} is disabled for security reasons
my $NO_PSQL_OPTION = 1;
## What type of output to use by default
my $DEFAULT_OUTPUT = 'nagios';
## Which user to connect as if --dbuser is not given
$opt{defaultuser} = 'postgres';
## If true, we show "after the pipe" statistics
$opt{showperf} = 1;
## If true, we show how long each query took by default. Requires Time::HiRes to be installed.
$opt{showtime} = 1;
## Default time display format, used for last_vacuum and last_analyze
my $SHOWTIME = 'HH24:MI FMMonth DD, YYYY';
## Always prepend 'postgres_' to the name of the service in the output string
my $FANCYNAME = 1;
## Change the service name to uppercase
my $YELLNAME = 1;
## Nothing below this line should need to be changed for normal usage.
## If you do find yourself needing to change something,
## please email the author as it probably indicates something
## that could be made into a command-line option or moved above.
## Messages are stored in these until the final output via finishup()
my (%ok, %warning, %critical, %unknown);
my $ME = basename($0);
my $ME2 = 'check_postgres.pl';
my $USAGE = qq{\nUsage: $ME <options>\n Try "$ME --help" for a complete list of options\n\n};
$opt{test} = 0;
$opt{timeout} = 10;
die $USAGE unless
GetOptions(
\%opt,
'version|V',
'verbose|v+',
'help|h',
'output=s',
'showperf=i',
'perflimit=i',
'showtime=i',
'timeout|t=i',
'test',
'action=s',
'warning=s',
'critical=s',
'include=s@',
'exclude=s@',
'includeuser=s@',
'excludeuser=s@',
'host|H=s@',
'port=s@',
'dbname|db=s@',
'dbuser|u=s@',
'dbpass=s@',
'PSQL=s',
'logfile=s', ## used by check_logfile only
'queryname=s', ## used by query_runtime only
)
and keys %opt
and ! @ARGV;
my $VERBOSE = $opt{verbose} || 0;
my $OUTPUT = $opt{output} || '';
## If not explicitly given an output, check the current directory,
## then fall back to the default.
if (!$OUTPUT) {
my $dir = getcwd;
if ($dir =~ /(nagios|munin)/io) {
$OUTPUT = lc $1;
}
else {
$OUTPUT = $DEFAULT_OUTPUT;
}
}
## Check for a valid output setting
if ($OUTPUT ne 'nagios' and $OUTPUT ne 'munin') { ## More coming soon...
die qq{Invalid output: must be 'nagios' or 'munin'\n};
}
## See if we need to invoke something based on our name
my $action = $opt{action} || '';
if ($ME =~ /check_postgres_(\w+)/) {
$action = $1;
}
$VERBOSE >= 3 and warn Dumper \%opt;
if ($opt{version}) {
print qq{$ME2 version $VERSION\n};
exit;
}
## Quick hash to put normal action information in one place:
my $action_info = {
# Name # clusterwide? # helpstring
backends => [1, 'Number of connections, compared to max_connections.'],
bloat => [0, 'Check for table and index bloat.'],
connection => [0, 'Simple connection check.'],
database_size => [0, 'Report if a database is too big.'],
disk_space => [1, 'Checks space of local disks Postgres is using.'],
index_size => [0, 'Checks the size of indexes only.'],
table_size => [0, 'Checks the size of tables only.'],
relation_size => [0, 'Checks the size of tables and indexes.'],
last_analyze => [0, 'Check the maximum time in seconds since any one table has been analyzed.'],
last_vacuum => [0, 'Check the maximum time in seconds since any one table has been vacuumed.'],
last_autoanalyze => [0, 'Check the maximum time in seconds since any one table has been autoanalyzed.'],
last_autovacuum => [0, 'Check the maximum time in seconds since any one table has been autovacuumed.'],
listener => [0, 'Checks for specific listeners.'],
locks => [0, 'Checks the number of locks.'],
logfile => [1, 'Checks that the logfile is being written to correctly.'],
query_runtime => [0, 'Check how long a specific query takes to run.'],
query_time => [1, 'Checks the maximum running time of current queries.'],
settings_checksum => [0, 'Check that no settings have changed since the last check.'],
timesync => [0, 'Compare database time to local system time.'],
txn_idle => [1, 'Checks the maximum "idle in transaction" time.'],
txn_time => [1, 'Checks the maximum open transaction time.'],
txn_wraparound => [1, 'See how close databases are getting to transaction ID wraparound.'],
version => [1, 'Check for proper Postgres version.'],
wal_files => [1, 'Check the number of WAL files in the pg_xlog directory'],
};
my $action_usage = '';
my $longname = 1;
for (keys %$action_info) {
$longname = length($_) if length($_) > $longname;
}
for (sort keys %$action_info) {
$action_usage .= sprintf " %-*s - %s\n", 2+$longname, $_, $action_info->{$_}[1];
}
if ($opt{help}) {
print qq{Usage: $ME2 <options>
Run various tests against one or more Postgres databases.
Returns with an exit code of 0 (success), 1 (warning), 2 (critical), or 3 (unknown)
This is version $VERSION.
Common connection options:
-H, --host=NAME hostname(s) to connect to; defaults to none (Unix socket)
-p, --port=NUM port(s) to connect to; defaults to 5432.
-db, --dbname=NAME database name(s) to connect to; defaults to 'postgres' or 'template1'
-u --dbuser=NAME database user(s) to connect as; defaults to 'postgres'
--dbpass=PASS database password(s); use a .pgpass file instead when possible
Connection options can be grouped: --host=a,b --host=c --port=1234 --port=3344
would connect to a-1234, b-1234, and c-3344
Limit options:
-w value, --warning=value the warning threshold, range depends on the action
-c value, --critical=value the critical threshold, range depends on the action
--include=name(s) items to specifically include (e.g. tables), depends on the action
--exclude=name(s) items to specifically exclude (e.g. tables), depends on the action
--includeuser=include objects owned by certain users
--excludeuser=exclude objects owned by certain users
Other options:
--PSQL=FILE location of the psql executable; avoid using if possible
-v, --verbose verbosity level; can be used more than once to increase the level
-h, --help display this help information
-t X, --timeout=X how long in seconds before we timeout. Defaults to 10 seconds.
Actions:
Which test is determined by the --action option, or by the name of the program
$action_usage
Special actions:
rebuild_symlinks - Make named symlinks to the main program for each action
rebuild_symlinks_force - Same as above, but removes existing symlinks first.
For a complete list of options and full documentation, please view the POD for this file.
Two ways to do this is to run:
pod2text $ME | less
pod2man $ME | man -l -
Or simply visit: http://bucardo.org/check_postgres/
};
exit;
}
$action =~ /\w/ or die $USAGE;
## Be nice and figure out what they meant
$action =~ s/\-/_/g;
$action = lc $action;
## Build symlinked copies of this file
build_symlinks() if $action =~ /build_symlinks/; ## Does not return, may be 'build_symlinks_force'
## Die if Time::HiRes is needed but not found
if ($opt{showtime}) {
eval {
require Time::HiRes; ## no critic
import Time::HiRes qw/gettimeofday tv_interval sleep/; ## no critic
};
if ($@) {
die qq{Cannot find Time::HiRes, needed if 'showtime' is true\n};
}
}
## We don't (usually) want to die, but want a graceful Nagios-like exit instead
sub ndie {
my $msg = shift;
chomp $msg;
print "ERROR: $msg\n";
exit 3;
}
## Everything from here on out needs psql, so find and verify a working version:
if ($NO_PSQL_OPTION) {
delete $opt{PSQL};
}
if (! defined $PSQL or ! length $PSQL) {
if (exists $opt{PSQL}) {
$PSQL = $opt{PSQL};
$PSQL =~ m{^/[\w\d\/]*psql$} or ndie qq{Invalid psql argument: must be full path to a file named psql\n};
-e $PSQL or ndie qq{Cannot find given psql executable: $PSQL\n};
}
else {
chomp($PSQL = qx{which psql});
$PSQL or ndie qq{Could not find a suitable psql executable\n};
}
}
-x $PSQL or ndie qq{The file "$PSQL" does not appear to be executable\n};
$res = qx{$PSQL --version};
$res =~ /^psql \(PostgreSQL\) (\d+\.\d+)/ or ndie qq{Could not determine psql version\n};
my $psql_version = $1;
$VERBOSE >= 1 and warn qq{psql=$PSQL version=$psql_version\n};
$opt{defaultdb} = $psql_version >= 7.4 ? 'postgres' : 'template1';
## Standard messages. Use these whenever possible when building actions.
my %template =
(
'T-EXCLUDE-DB' => 'No matching databases found due to exclusion/inclusion options',
'T-EXCLUDE-FS' => 'No matching file systems found due to exclusion/inclusion options',
'T-EXCLUDE-REL' => 'No matching relations found due to exclusion/inclusion options',
'T-EXCLUDE-SET' => 'No matching settings found due to exclusion/inclusion options',
'T-EXCLUDE-TABLE' => 'No matching tables found due to exclusion/inclusion options',
'T-EXCLUDE-USEROK' => 'No matching entries found due to user exclusion/inclusion options',
'T-BAD-QUERY' => 'Invalid query returned:',
);
sub add_response {
my ($type,$msg) = @_;
my $header = sprintf q{%s%s%s},
$action_info->{$action}[0] ? '' : qq{DB "$db->{dbname}" },
$db->{host} eq '<none>' ? '' : qq{(host:$db->{host}) },
$db->{port} eq '5432' ? '' : qq{(port=$db->{port}) };
$header =~ s/\s+$//;
my $perf = $opt{showtime} ? "time=$db->{totaltime}" : '';
if ($db->{perf}) {
$perf .= " $db->{perf}";
}
$msg =~ s/(T-[\w\-]+)/$template{$1}/g;
push @{$type->{$header}} => [$msg,$perf];
}
sub add_unknown { ## no critic
my $msg = shift || $db->{error};
add_response \%unknown, $msg;
}
sub add_critical {
add_response \%critical, shift;
}
sub add_warning {
add_response \%warning, shift;
}
sub add_ok {
add_response \%ok, shift;
}
sub finishup {
## Final output
## These are meant to be compact and terse: sometimes messages go to pagers
$action =~ s/^\s*(\S+)\s*$/$1/;
my $service = sprintf "%s$action", $FANCYNAME ? 'postgres_' : '';
if (keys %critical or keys %warning or keys %ok or keys %unknown) {
printf '%s ', $YELLNAME ? uc $service : $service;
}
sub dumpresult {
my $SEP = ' * ';
my $type = shift;
for (sort keys %$type) {
printf "$_ %s ", join $SEP => map { $_->[0] } @{$type->{$_}};
}
if ($opt{showperf}) {
print '| ';
for (sort keys %$type) {
printf '%s ', join $SEP => map { $_->[1] } @{$type->{$_}};
}
}
print "\n";
}
if (keys %critical) {
print 'CRITICAL: ';
dumpresult(\%critical);
exit 2;
}
if (keys %warning) {
print 'WARNING: ';
dumpresult(\%warning);
exit 1;
}
if (keys %ok) {
print 'OK: ';
dumpresult(\%ok);
exit 0;
}
if (keys %unknown) {
print 'UNKNOWN: ';
dumpresult(\%unknown);
exit 3;
}
die $USAGE;
} ## end of finishup
## For options that take a size e.g. --critical="10 GB"
my $sizere = qr{^\s*(\d+\.?\d?)\s*([bkmgtpz])?\w*$}i; ## Don't care about the rest of the string
## For options that take a time e.g. --critical="10 minutes" Fractions are allowed.
my $timere = qr{^\s*(\d+(?:\.\d+)?)\s*(\w*)\s*$}i;
## For options that must be specified in seconds
my $timesecre = qr{^\s*(\d+)\s*(?:s(?:econd|ec)?)?s?\s*$};
## For simple checksums:
my $checksumre = qr{^[a-f0-9]{32}$};
## If in test mode, verify that we can run each requested action
my %testaction = (
last_vacuum => 'ON: stats_row_level VERSION: 8.2',
last_analyze => 'ON: stats_row_level VERSION: 8.2',
last_autovacuum => 'ON: stats_row_level VERSION: 8.2',
last_autoanalyze => 'ON: stats_row_level VERSION: 8.2',
database_size => 'VERSION: 8.1',
relation_size => 'VERSION: 8.1',
table_size => 'VERSION: 8.1',
index_size => 'VERSION: 8.1',
txn_idle => 'VERSION: 8.2',
txn_time => 'VERSION: 8.3',
wal_files => 'VERSION: 8.1',
);
if ($opt{test}) {
print "BEGIN TEST MODE\n";
my $info = run_command('SELECT name, setting FROM pg_settings');
my %set; ## port, host, name, user
for my $db (@{$info->{db}}) {
if (exists $db->{fail}) {
(my $err = $db->{error}) =~ s/\s*\n\s*/ \| /g;
print "Connection failed: $db->{pname} $err\n";
next;
}
print "Connection ok: $db->{pname}\n";
for (split /\n/ => $db->{slurp}) {
while (/(\S+)\s*\|\s*(.+)\s*/sg) {
$set{$db->{pname}}{$1} = $2;
}
}
}
for my $ac (split /\s+/ => $action) {
my $limit = $testaction{lc $ac};
next if ! defined $limit;
while ($limit =~ /\bON: (\w+)/g) {
my $setting = $1;
for my $db (@{$info->{db}}) {
next unless exists $db->{ok};
my $val = $set{$db->{pname}}{$setting};
if ($val ne 'on') {
print qq{Cannot run "$ac" on $db->{pname}: $setting is not set to on\n};
}
}
}
if ($limit =~ /VERSION: ((\d+)\.(\d+))/) {
my ($rver,$rmaj,$rmin) = ($1,$2,$3);
for my $db (@{$info->{db}}) {
next unless exists $db->{ok};
if ($set{$db->{pname}}{server_version} !~ /((\d+)\.(\d+))/) {
print "Could not find version for $db->{pname}\n";
next;
}
my ($sver,$smaj,$smin) = ($1,$2,$3);
if ($smaj < $rmaj or ($smaj==$rmaj and $smin < $rmin)) {
print qq{Cannot run "$ac" on $db->{pname}: version must be >= $rver, but is $sver\n};
}
}
}
}
print "END OF TEST MODE\n";
exit;
}
## Expand the list of included/excluded users into a standard format
my $USERWHERECLAUSE = '';
if ($opt{includeuser}) {
my %userlist;
for my $user (@{$opt{includeuser}}) {
for my $u2 (split /,/ => $user) {
$userlist{$u2}++;
}
}
my $safename;
if (1 == keys %userlist) {
($safename = each %userlist) =~ s/'/''/g;
$USERWHERECLAUSE = " AND usename = '$safename'";
}
else {
$USERWHERECLAUSE = ' AND usename IN (';
for my $user (sort keys %userlist) {
($safename = $user) =~ s/'/''/g;
$USERWHERECLAUSE .= "'$safename',";
}
chop $USERWHERECLAUSE;
$USERWHERECLAUSE .= ')';
}
}
elsif ($opt{excludeuser}) {
my %userlist;
for my $user (@{$opt{excludeuser}}) {
for my $u2 (split /,/ => $user) {
$userlist{$u2}++;
}
}
my $safename;
if (1 == keys %userlist) {
($safename = each %userlist) =~ s/'/''/g;
$USERWHERECLAUSE = " AND usename <> '$safename'";
}
else {
$USERWHERECLAUSE = ' AND usename NOT IN (';
for my $user (sort keys %userlist) {
($safename = $user) =~ s/'/''/g;
$USERWHERECLAUSE .= "'$safename',";
}
chop $USERWHERECLAUSE;
$USERWHERECLAUSE .= ')';
}
}
## Check number of connections, compare to max_connections
check_backends() if $action eq 'backends';
## Table and index bloat
check_bloat() if $action eq 'bloat';
## Simple connection, warning or critical options
check_connection() if $action eq 'connection';
## Check the size of one or more databases
check_database_size() if $action eq 'database_size';
## Check local disk_space - local means it must be run from the same box!
check_disk_space() if $action eq 'disk_space';
## Check the size of relations, or more specifically, tables and indexes
check_index_size() if $action eq 'index_size';
check_table_size() if $action eq 'table_size';
check_relation_size() if $action eq 'relation_size';
## Check how long since the last full analyze
check_last_analyze() if $action eq 'last_analyze';
## Check how long since the last full vacuum
check_last_vacuum() if $action eq 'last_vacuum';
## Check how long since the last AUTOanalyze
check_last_analyze('auto') if $action eq 'last_autoanalyze';
## Check how long since the last full AUTOvacuum
check_last_vacuum('auto') if $action eq 'last_autovacuum';
## Check that someone is listening for a specific thing
check_listener() if $action eq 'listener';
## Check number and type of locks
check_locks() if $action eq 'locks';
## Logfile is being written to
check_logfile() if $action eq 'logfile';
## Known query finishes in a good amount of time
check_query_runtime() if $action eq 'query_runtime';
## Check the length of running queries
check_query_time() if $action eq 'query_time';
## Verify that the settings are what we think they should be
check_settings_checksum() if $action eq 'settings_checksum';
## Compare DB time to localtime, alert on number of seconds difference
check_timesync() if $action eq 'timesync';
## Check for transaction ID wraparound in all databases
check_txn_wraparound() if $action eq 'txn_wraparound';
## Compare DB versions. warning = just major.minor, critical = full string
check_version() if $action eq 'version';
## Check the number of WAL files. warning and critical are numbers
check_wal_files() if $action eq 'wal_files';
## Check the maximum transaction age of all connections
check_txn_time() if $action eq 'txn_time';
## Check the maximum age of idle in transaction connections
check_txn_idle() if $action eq 'txn_idle';
finishup();
exit;
sub build_symlinks {
## Create symlinks to most actions
$ME =~ /postgres/
or die qq{This command will not work unless the program has the word "postgres" in it\n};
my $force = $action =~ /force/ ? 1 : 0;
for my $action (sort keys %$action_info) {
my $space = ' ' x ($longname - length $action);
my $file = "check_postgres_$action";
if (-l $file) {
if (!$force) {
my $source = readlink $file;
print qq{Not creating "$file":$space already linked to "$source"\n};
next;
}
print qq{Unlinking "$file":$space };
unlink $file or die qq{Failed to unlink "$file": $!\n};
}
elsif (-e $file) {
print qq{Not creating "$file":$space file already exists\n};
next;
}
if (symlink $0, $file) {
print qq{Created "$file"\n};
}
else {
print qq{Could not symlink $file to $ME: $!\n};
}
}
exit;
} ## end of build_symlinks
sub pretty_size {
## Transform number of bytes to a SI display similar to Postgres' format
my $bytes = shift;
return "$bytes bytes" if $bytes < 10240;
my @unit = qw/kB MB GB TB PB EB YB ZB/;
for my $p (1..@unit) {
if ($bytes <= 1024**$p) {
$bytes /= (1024**($p-1));
return sprintf '%.2f %s', $bytes, $unit[$p-2];
}
}
return $bytes;
} ## end of pretty_size
sub run_command {
## Run a command string against each of our databases using psql
## Optional args in a hashref:
## "failok" - don't report if we failed
## "target" - use this targetlist instead of generating one
## "timeout" - change the timeout from the default of $opt{timeout}
## "regex" - the query must match this or we throw an error
## "emptyok" - it's okay to not match any rows at all
my $string = shift;
my $arg = shift || {};
my $info = { command => $string, db => [], hosts => 0 };
$VERBOSE >= 3 and warn qq{Starting run_command with "$string"\n};
my (%host,$passfile,$passfh,$tempdir,$tempfile,$tempfh,$errorfile,$errfh);
my $offset = -1;
## Build a list of all databases to connect to.
## Number is determined by host, port, and db arguments
## Multi-args are grouped together: host, port, dbuser, dbpass
## Grouped are kept together for first pass
## The final arg in a group is passed on
##
## Examples:
## --host=a,b --port=5433 --db=c
## Connects twice to port 5433, using database c, to hosts a and b
## a-5433-c b-5433-c
##
## --host=a,b --port=5433 --db=c,d
## Connects four times: a-5433-c a-5433-d b-5433-c b-5433-d
##
## --host=a,b --host=foo --port=1234 --port=5433 --db=e,f
## Connects six times: a-1234-e a-1234-f b-1234-e b-1234-f foo-5433-e foo-5433-f
##
## --host=a,b --host=x --port=5432,5433 --dbuser=alice --dbuser=bob -db=baz
## Connects three times: a-5432-alice-baz b-5433-alice-baz x-5433-bob-baz
## The final list of targets:
my @target;
## Default connection options
my $conn =
{
host => ['<none>'],
port => [5432],
dbname => [$opt{defaultdb}],
dbuser => [$opt{defaultuser}],
dbpass => [''],
};
my $gbin = 0;
GROUP: {
## This level controls a "group" of targets
## If we were passed in a target, use that and move on
if (exists $arg->{target}) {
push @target, $arg->{target};
last GROUP;
}
my %group;
my $foundgroup = 0;
for my $v (keys %$conn) {
## Something new?
if (defined $opt{$v}->[$gbin]) {
my $new = $opt{$v}->[$gbin];
$new =~ s/\s+//g;
## Set this as the new default
$conn->{$v} = [split /,/ => $new];
$foundgroup = 1;
}
$group{$v} = $conn->{$v};
}
if (!$foundgroup) { ## Nothing new, so we bail
last GROUP;
}
$gbin++;
## Now break the newly created group into individual targets
my $tbin = 0;
TARGET: {
my $foundtarget = 0;
## We know th
my %temptarget;
for my $g (keys %group) {
if (defined $group{$g}->[$tbin]) {
$conn->{$g} = [$group{$g}->[$tbin]];
$foundtarget = 1;
}
$temptarget{$g} = $conn->{$g}[0];
}
## Leave if nothing new
last TARGET if ! $foundtarget;
## Add to our master list
push @target, \%temptarget;
$tbin++;
redo;
} ## end TARGET
redo;
} ## end GROUP
if (! @target) {
ndie qq{No target databases found\n};
}
## Create a temp file to store our results
$tempdir = tempdir(CLEANUP => 1);
($tempfh,$tempfile) = tempfile('nagios_psql.XXXXXXX', SUFFIX => '.tmp', DIR => $tempdir);
## Create another one to catch any errors
($errfh,$errorfile) = tempfile('nagios_psql_stderr.XXXXXXX', SUFFIX => '.tmp', DIR => $tempdir);
for $db (@target) {
## Just to keep things clean:
truncate $tempfh, 0;
truncate $errfh, 0;
## Store this target in the global target list
push @{$info->{db}}, $db;
$db->{pname} = "port=$db->{port} host=$db->{host} db=$db->{dbname} user=$db->{dbuser}";
my @args = ('-q', '-U', "$db->{dbuser}", '-d', $db->{dbname}, '-t');
if ($db->{host} ne '<none>') {
push @args => '-h', $db->{host};
$host{$db->{host}}++; ## For the overall count
}
push @args => '-p', $db->{port};
if (defined $db->{dbpass} and length $db->{dbpass}) {
## Make a custom PGPASSFILE. Far better to simply use your own .pgpass of course
($passfh,$passfile) = tempfile('nagios.XXXXXXXX', SUFFIX => '.tmp', DIR => $tempdir);
$VERBOSE >= 3 and warn "Created temporary pgpass file $passfile\n";
$ENV{PGPASSFILE} = $passfile;
printf $passfh "%s:%s:%s:%s:%s\n",
$db->{host} eq '<none>' ? '*' : $db->{host}, $db->{port}, $db->{dbname}, $db->{dbuser}, $db->{dbpass};
close $passfh or ndie qq{Could not close $passfile: $!\n};
}
push @args, '-o', $tempfile;
push @args, '-c', $string;
$VERBOSE >= 3 and warn Dumper \@args;
local $SIG{ALRM} = sub { die 'Timed out' };
my $timeout = $arg->{timeout} || $opt{timeout};
alarm 0;
my $start = $opt{showtime} ? [gettimeofday()] : 0; ## no critic
open my $oldstderr, '>&', STDERR or ndie "Could not dupe STDERR\n";
open STDERR, '>', $errorfile or ndie qq{Could not open STDERR?!\n};
eval {
alarm $timeout;
$res = system $PSQL => @args;
};
my $err = $@;
alarm 0;
open STDERR, '>&', $oldstderr or ndie "Could not recreate STDERR\n";
close $oldstderr or ndie qq{Could not close STDERR copy: $!\n};
if ($err) {
if ($err =~ /Timed out/) {
ndie qq{Command timed out! Consider boosting --timeout higher than $timeout\n};
}
else {
ndie q{Unknown error inside of the "run_command" function};
}
}
$db->{totaltime} = sprintf '%.2f', $opt{showtime} ? tv_interval($start) : 0; ## no critic
if ($res) {
$db->{fail} = $res;
$VERBOSE >= 3 and !$arg->{failok} and warn qq{System call failed with a $res\n};
seek $errfh, 0, 0;
{
local $/;
$db->{error} = <$errfh> || '';
$db->{error} =~ s/\s*$//;
$db->{error} =~ s/^psql: //;
}
if (!$db->{ok} and !$arg->{failok}) {
add_unknown;
## Remove it from the returned hash
pop @{$info->{db}};
}
}
else {
seek $tempfh, 0, 0;
{
local $/;
$db->{slurp} = <$tempfh>;
}
$db->{ok} = 1;
## Allow an empty query (no matching rows) if requested
if ($arg->{emptyok} and $db->{slurp} =~ /^\s*$/o) {
}
## If we were provided with a regex, check and bail if it fails
elsif ($arg->{regex}) {
if ($db->{slurp} !~ $arg->{regex}) {
add_unknown qq{T-BAD-QUERY $db->{slurp}};
## Remove it from the returned hash
pop @{$info->{db}};
}
}
}
} ## end each database
close $errfh or ndie qq{Could not close $errorfile: $!\n};
close $tempfh or ndie qq{Could not close $tempfile: $!\n};
$info->{hosts} = keys %host;
$VERBOSE >= 3 and warn Dumper $info;
return $info;
} ## end of run_command
sub size_in_bytes { ## no critic (RequireArgUnpacking)
## Given a number and a unit, return the number of bytes.
my ($val,$unit) = ($_[0],lc substr($_[1]||'s',0,1));
return $val * ($unit eq 's' ? 1 : $unit eq 'k' ? 1024 : $unit eq 'm' ? 1024**2 :
$unit eq 'g' ? 1024**3 : $unit eq 't' ? 1024**4 :
$unit eq 'p' ? 1024**5 : $unit eq 'e' ? 1024**6 :
$unit eq 'z' ? 1024**7 : 1024**8);
} ## end of size_in_bytes
sub size_in_seconds {
my ($string,$type) = @_;
return '' if ! length $string;
if ($string !~ $timere) {
my $l = substr($type,0,1);
ndie qq{Value for '$type' must be a valid time. Examples: -$l 1s -$l "10 minutes"\n};
}
my ($val,$unit) = ($1,lc substr($2||'s',0,1));
my $tempval = sprintf '%.9f', $val * ($unit eq 's' ? 1 : $unit eq 'm' ? 60 : $unit eq 'h' ? 3600 : 86600);
$tempval =~ s/0+$//;
$tempval = int $tempval if $tempval =~ /\.$/;
return $tempval;
} ## end of size_in_seconds
sub skip_item {
## Determine if something should be skipped due to inclusion/exclusion options
## Exclusion checked first: inclusion can pull it back out.
my $name = shift;
my $stat = 0;
## Is this excluded?
if (defined $opt{exclude}) {
$stat = 1;
for (@{$opt{exclude}}) {
for my $ex (split /\s*,\s*/ => $_) {
if ($ex =~ s/^~//) {
($stat += 2 and last) if $name =~ /$ex/;
}
else {
($stat += 2 and last) if $name eq $ex;
}
}
}
}
if (defined $opt{include}) {
$stat += 4;
for (@{$opt{include}}) {
for my $in (split /\s*,\s*/ => $_) {
if ($in =~ s/^~//) {
($stat += 8 and last) if $name =~ /$in/;
}
else {
($stat += 8 and last) if $name eq $in;
}
}
}
}
## Easiest to state the cases when we DO skip:
return 1 if
3 == $stat ## exclude matched, no inclusion checking
or 4 == $stat ## include check only, no match
or 7 == $stat; ## exclude match, no inclusion match
return 0;
} ## end of skip_item
sub validate_range {
## Valid that warning and critical are set correctly.
## Returns new values of both
my $arg = shift;
defined $arg and ref $arg eq 'HASH' or ndie qq{validate_range must be called with a hashref\n};
my $type = $arg->{type} or ndie qq{validate_range must be provided a 'type'\n};
## The 'default default' is an empty string, which should fail all mandatory tests
## We only set the 'arg' default if neither option is provided.
my $warning = exists $opt{warning} ? $opt{warning} :
exists $opt{critical} ? '' : $arg->{default_warning} || '';
my $critical = exists $opt{critical} ? $opt{critical} :
exists $opt{warning} ? '' : $arg->{default_critical} || '';
if ('seconds' eq $type) {
if (length $warning) {
if ($warning !~ $timesecre) {
ndie qq{Invalid argument to 'warning' option: must be number of seconds\n};
}
$warning = $1;
}
if (length $critical) {
if ($critical !~ $timesecre) {
ndie qq{Invalid argument to 'critical' option: must be number of seconds\n};
}
$critical = $1;
if (length $warning and $warning > $critical) {
ndie qq{The 'warning' option ($warning s) cannot be larger than the 'critical' option ($critical s)\n};
}
}
}
elsif ('time' eq $type) {
$critical = size_in_seconds($critical, 'critical');
$warning = size_in_seconds($warning, 'warning');
if (! length $critical and ! length $warning) {
ndie qq{Must provide a warning and/or critical time\n};
}
if (length $warning and length $critical and $warning > $critical) {
ndie qq{The 'warning' option ($warning s) cannot be larger than the 'critical' option ($critical s)\n};
}
}
elsif ('version' eq $type) {
my $msg = q{must be in the format X.Y or X.Y.Z, where X is the major version number, }
.q{Y is the minor version number, and Z is the revision};
if (length $warning and $warning !~ /^\d+\.\d\.?[\d\w]*$/) {
ndie qq{Invalid string for 'warning' option: $msg};
}
if (length $critical and $critical !~ /^\d+\.\d\.?[\d\w]*$/) {
ndie qq{Invalid string for 'critical' option: $msg};
}
if (! length $critical and ! length $warning) {
ndie "Must provide a 'warning' option, a 'critical' option, or both\n";
}
}
elsif ('size' eq $type) {
if (length $critical) {
if ($critical !~ $sizere) {
ndie "Invalid size for 'critical' option\n";
}
$critical = size_in_bytes($1,$2);
}
if (length $warning) {
if ($warning !~ $sizere) {
ndie "Invalid size for 'warning' option\n";
}
$warning = size_in_bytes($1,$2);
if (length $critical and $warning > $critical) {
ndie qq{The 'warning' option ($warning bytes) cannot be larger than the 'critical' option ($critical bytes)\n};
}
}
elsif (!length $critical) {
ndie qq{Must provide a warning and/or critical size\n};
}
}
elsif ($type =~ /integer/) {
$warning =~ s/_//g;
if (length $warning and $warning !~ /^\d+$/) {
ndie sprintf "Invalid argument for 'warning' option: must be %s integer\n",
$type =~ /positive/ ? 'a positive' : 'an';
}
$critical =~ s/_//g;
if (length $critical and $critical !~ /^\d+$/) {
ndie sprintf "Invalid argument for 'critical' option: must be %s integer\n",
$type =~ /positive/ ? 'a positive' : 'an';
}
if (length $warning and length $critical and $warning > $critical) {
ndie qq{The 'warning' option cannot be greater than the 'critical' option\n};
}
}
elsif ('restringex' eq $type) {
if (! length $critical and ! length $warning) {
ndie qq{Must provide a 'warning' or 'critical' option\n};
}
if (length $critical and length $warning) {
ndie qq{Can only provide 'warning' OR 'critical' option\n};
}
my $string = length $critical ? $critical : $warning;
my $regex = ($string =~ s/^~//) ? '~' : '=';
$string =~ /^\w+$/ or die qq{Invalid option\n};
}
elsif ('size or percent' eq $type) {
if (length $critical) {
if ($critical =~ $sizere) {
$critical = size_in_bytes($1,$2);
}
elsif ($critical !~ /^\d\d?\%$/) {
ndie qq{Invalid 'critical' option: must be size or percentage\n};
}
}
if (length $warning) {
if ($warning =~ $sizere) {
$warning = size_in_bytes($1,$2);
}
elsif ($warning !~ /^\d\d?\%$/) {
ndie qq{Invalid 'warning' option: must be size or percentage\n};
}
}
elsif (! length $critical) {
ndie qq{Must provide a warning and/or critical size\n};
}
}
elsif ('checksum' eq $type) {
if (length $critical and $critical !~ $checksumre and $critical ne '0') {
ndie qq{Invalid 'critical' option: must be a checksum\n};
}
if (length $warning and $warning !~ $checksumre) {
ndie qq{Invalid 'warning' option: must be a checksum\n};
}
}
elsif ('multival' eq $type) { ## Simple number, or foo=#;bar=#
my %err;
while ($critical =~ /(\w+)\s*=\s*(\d+)/gi) {
my ($name,$val) = (lc $1,$2);
$name =~ s/lock$//;
$err{$name} = $val;
}
if (keys %err) {
$critical = \%err;
}
elsif (length $critical and $critical !~ /^\d+$/) {
ndie qq{Invalid 'critical' option: must be number of locks, or "type1=#;type2=#"\n};
}
my %warn;
while ($warning =~ /(\w+)\s*=\s*(\d+)/gi) {
my ($name,$val) = (lc $1,$2);
$name =~ s/lock$//;
$warn{$name} = $val;
}
if (keys %warn) {
$warning = \%warn;
}
elsif (length $warning and $warning !~ /^\d+$/) {
ndie qq{Invalid 'warning' option: must be number of locks, or "type1=#;type2=#"\n};
}
}
else {
ndie qq{validate_range called with unknown type '$type'\n};
}
if ($arg->{both}) {
if (! length $warning or ! length $critical) {
ndie qq{Must provide both 'warning' and 'critical' options\n};
}
}
if ($arg->{leastone}) {
if (! length $warning and ! length $critical) {
ndie qq{Must provide at least a 'warning' or 'critical' option\n};
}
}
elsif ($arg->{onlyone}) {
if (length $warning and length $critical) {
ndie qq{Can only provide 'warning' OR 'critical' option\n};
}
if (! length $warning and ! length $critical) {
ndie qq{Must provide either 'critical' or 'warning' option\n};
}
}
return ($warning,$critical);
} ## end of validate_range
sub check_backends {
## Check the number of connections
## It makes no sense to run this more than once on the same cluster
## Need to be superuser, else only your queries will be visible
## Warning and criticals can take three forms:
## critical = 12 -- complain if there are 12 or more connections
## critical = 95% -- complain if >= 95% of available connections are used
## critical = -5 -- complain if there are only 5 or fewer connection slots left
## Can also ignore databases with exclude, and limit with include
## The former two options only work with simple numbers - no percentage or negative
my $warning = $opt{warning} || '90%';
my $critical = $opt{critical} || '95%';
my $validre = qr{^(\-?)(\d+)(\%?)$};
if ($warning !~ $validre) {
ndie "Warning for number of users must be a number or percentage\n";
}
my ($w1,$w2,$w3) = ($1,$2,$3);
if ($critical !~ $validre) {
ndie "Critical for number of users must be a number or percentage\n";
}
my ($e1,$e2,$e3) = ($1,$2,$3);
if ($w2 > $e2 and $w1 eq $e1 and $w3 eq $e3 and $w1 eq '') {
ndie qq{Makes no sense for warning to be greater than critical!\n};
}
if ($w2 < $e2 and $w1 eq $e1 and $w3 eq $e3 and $w1 eq '-') {
ndie qq{Makes no sense for warning to be less than critical!\n};
}
if (($w1 and $w3) or ($e1 and $e3)) {
ndie qq{Cannot specify a negative percent!\n};
}
$SQL = q{SELECT setting FROM pg_settings WHERE name = 'max_connections'};
$SQL = "SELECT COUNT(*), ($SQL), datname FROM pg_stat_activity GROUP BY 2,3";
my $info = run_command($SQL, {regex => qr[\s*\d+ \| \d+\s+\|] } );
for $db (@{$info->{db}}) {
my ($limit,$total) = 0;
SLURP: while ($db->{slurp} =~ /(\d+) \| (\d+)\s+\|\s+(\w+)\s*/gsm) {
$limit ||= $2;
my ($current,$dbname) = ($1,$3);
next SLURP if skip_item($dbname);
$db->{perf} .= " $dbname=$current";
$total += $current;
}
if (!$total) {
add_unknown 'T-EXCLUDE-DB';
next;
}
my $msg = qq{$total of $limit connections};
my $ok = 1;
if ($e1) { ## minus
$ok = 0 if $limit-$total >= $e2;
}
elsif ($e3) { ## percent
my $nowpercent = $total/$limit*100;
$ok = 0 if $nowpercent >= $e2;
}
else { ## raw number
$ok = 0 if $total >= $e2;
}
if (!$ok) {
add_critical $msg;
next;
}
if ($w1) {
$ok = 0 if $limit-$total >= $w2;
}
elsif ($w3) {
my $nowpercent = $total/$limit*100;
$ok = 0 if $nowpercent >= $w2;
}
else {
$ok = 0 if $total >= $w2;
}
if (!$ok) {
add_warning $msg;
next;
}
add_ok $msg;
}
return;
} ## end of check_backends
sub check_bloat {
## Check how bloated the tables and indexes are
## NOTE! This check depends on ANALYZE being run regularly
## Also requires stats collection to be on
## This action may be very slow on large databases
## By default, checks all relations
## Can check specific one(s) with include; can ignore some with exclude
## Begin name with a '~' to make it a regular expression
## Warning and critical are in sizes, defaults to bytes
## Valid units: b, k, m, g, t, e
## All above may be written as plural or with a trailing 'b'
## Example: --critical="25 GB" --include="mylargetable"
## Don't bother with tables or indexes unless they have at least this many bloated pages
my $MINPAGES = 0;
my $MINIPAGES = 10;
my $LIMIT = 10;
if ($opt{perflimit}) {
$LIMIT = $opt{perflimit};
}
my ($warning, $critical) = validate_range
({
type => 'size',
default_warning => '1 GB',
default_critical => '5 GB',
});
## This was fun to write
$SQL = qq{
SELECT
schemaname, tablename, reltuples::bigint, relpages::bigint, otta,
ROUND(CASE WHEN otta=0 THEN 0.0 ELSE sml.relpages/otta::numeric END,1) AS tbloat,
relpages::bigint - otta AS wastedpages,
bs*(sml.relpages-otta)::bigint AS wastedbytes,
pg_size_pretty((bs*(relpages-otta))::bigint) AS wastedsize,
iname, ituples::bigint, ipages::bigint, iotta,
ROUND(CASE WHEN iotta=0 OR ipages=0 THEN 0.0 ELSE ipages/iotta::numeric END,1) AS ibloat,
CASE WHEN ipages < iotta THEN 0 ELSE ipages::bigint - iotta END AS wastedipages,
CASE WHEN ipages < iotta THEN 0 ELSE bs*(ipages-iotta) END AS wastedibytes,
CASE WHEN ipages < iotta THEN pg_size_pretty(0) ELSE pg_size_pretty((bs*(ipages-iotta))::bigint) END AS wastedisize
FROM (
SELECT
schemaname, tablename, cc.reltuples, cc.relpages, bs,
CEIL((cc.reltuples*((datahdr+ma-
(CASE WHEN datahdr%ma=0 THEN ma ELSE datahdr%ma END))+nullhdr2+4))/(bs-20::float)) AS otta,
COALESCE(c2.relname,'?') AS iname, COALESCE(c2.reltuples,0) AS ituples, COALESCE(c2.relpages,0) AS ipages,
COALESCE(CEIL((c2.reltuples*(datahdr-12))/(bs-20::float)),0) AS iotta -- very rough approximation, assumes all cols
FROM (
SELECT
ma,bs,schemaname,tablename,
(datawidth+(hdr+ma-(case when hdr%ma=0 THEN ma ELSE hdr%ma END)))::numeric AS datahdr,
(maxfracsum*(nullhdr+ma-(case when nullhdr%ma=0 THEN ma ELSE nullhdr%ma END))) AS nullhdr2
FROM (
SELECT
schemaname, tablename, hdr, ma, bs,
SUM((1-null_frac)*avg_width) AS datawidth,
MAX(null_frac) AS maxfracsum,
hdr+(
SELECT 1+count(*)/8
FROM pg_stats s2
WHERE null_frac<>0 AND s2.schemaname = s.schemaname AND s2.tablename = s.tablename
) AS nullhdr
FROM pg_stats s, (
SELECT
(SELECT current_setting('block_size')::numeric) AS bs,
CASE WHEN substring(v,12,3) IN ('8.0','8.1','8.2') THEN 27 ELSE 23 END AS hdr,
CASE WHEN v ~ 'mingw32' THEN 8 ELSE 4 END AS ma
FROM (SELECT version() AS v) AS foo
) AS constants
GROUP BY 1,2,3,4,5
) AS foo
) AS rs
JOIN pg_class cc ON cc.relname = rs.tablename
JOIN pg_namespace nn ON cc.relnamespace = nn.oid AND nn.nspname = rs.schemaname
LEFT JOIN pg_index i ON indrelid = cc.oid
LEFT JOIN pg_class c2 ON c2.oid = i.indexrelid
) AS sml
WHERE sml.relpages - otta > $MINPAGES OR ipages - iotta > $MINIPAGES
ORDER BY wastedbytes DESC LIMIT $LIMIT
};
my $info = run_command($SQL);
## schema, table, rows, pages, otta, bloat, wastedpages, wastedbytes, wastedsize
## index, "" "" ...
my $N = qr{ (.+?)\s*\|};
my $D = qr{\s+(\d+) \|};
my $F = qr{\s+(\d+\.\d) \|};
my $S = qr{ (\d+ \w+)\s+\|};
my $E = qr{ (\d+ \w+)\s*};
my $L = qr{$N$N$D$D$D$F$D$D$S$N$D$D$D$F$D$D$E$};
my %seenit;
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /$L/) {
add_ok q{no relations meet the minimum bloat criteria};
next;
}
## Not a 'regex' to run_command as we need to check the above first.
if ($db->{slurp} !~ /\d+\s*\| \d+/) {
add_unknown qq{T-BAD-QUERY $db->{slurp}};
next;
}
my $max = -1;
my $maxmsg;
SLURP: while ($db->{slurp} =~ /$L/gsm) {
my ($schema,$table,$tups,$pages,$otta,$bloat,$wp,$wb,$ws,
$index,$irows,$ipages,$iotta,$ibloat,$iwp,$iwb,$iws)
= ($1,$2,$3,$4,$5,$6,$7,$8,$9,$10,$11,$12,$13,$14,$15,$16,$17,$18);
next SLURP if skip_item($table);
## Made it past the exclusions
$max = -2 if $max == -1;
## Do the table first if we haven't seen it
if (! $seenit{"$schema.$table"}++) {
$db->{perf} .= " $schema.$table=$wb";
my $msg = qq{table $schema.$table rows:$tups pages:$pages shouldbe:$otta (${bloat}X)};
$msg .= qq{ wasted size:$wb ($ws)};
## The key here is the wastedbytes
if ($wb >= $critical) {
add_critical $msg;
}
elsif ($wb >= $warning) {
add_warning $msg;
}
else {
($max = $wb, $maxmsg = $msg) if $wb > $max;
}
}
## Now the index, if it exists
if ($index ne '?') {
$db->{perf} .= " $index=$iwb" if $iwb;
my $msg = qq{index '$index' rows:$irows pages:$ipages shouldbe:$iotta (${ibloat}X)};
$msg .= qq{ wasted bytes:$iwb ($iws)};
if ($iwb >= $critical) {
add_critical $msg;
}
elsif ($iwb >= $warning) {
add_warning $msg;
}
else {
($max = $iwb, $maxmsg = $msg) if $iwb > $max;
}
}
}
if ($max == -1) {
add_unknown 'T-EXCLUDE-REL';
}
elsif ($max != -1) {
add_ok $maxmsg;
}
}
return;
} ## end of check_bloat
sub check_connection {
## Check the connection, get the connection time and version
## No comparisons made: warning and critical are not allowed
if ($opt{warning} or $opt{critical}) {
ndie qq{No warning or critical options are needed\n};
}
my $info = run_command('SELECT version()');
## Parse it out and return our information
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /PostgreSQL (\S+)/o) {
add_unknown "T-BAD-QUERY $db->{slurp}";
next;
}
add_ok "version $1";
}
return;
} ## end of check_connection
sub check_database_size {
## Check the size of one or more databases
## By default, checks all databases
## Can check specific one(s) with include
## Can ignore some with exclude
## Warning and critical are bytes
## Valid units: b, k, m, g, t, e
## All above may be written as plural or with a trailing 'b'
## Limit to a specific user (db owner) with the includeuser option
## Exclude users with the excludeuser option
my ($warning, $critical) = validate_range({type => 'size'});
$USERWHERECLAUSE =~ s/AND/WHERE/;
$SQL = q{SELECT pg_database_size(d.oid), pg_size_pretty(pg_database_size(d.oid)), datname, usename }.
qq{FROM pg_database d JOIN pg_user u ON (u.usesysid=d.datdba)$USERWHERECLAUSE};
if ($opt{perflimit}) {
$SQL .= " ORDER BY 1 DESC LIMIT $opt{perflimit}";
}
my $info = run_command($SQL, {regex => qr{\d+ \|}, emptyok => 1 } );
for $db (@{$info->{db}}) {
my $max = -1;
my %s;
SLURP: while ($db->{slurp} =~ /(\d+) \| (\d+ \w+)\s+\| (\S+)/gsm) {
my ($size,$psize,$name) = ($1,$2,$3);
next SLURP if skip_item($name);
$max=$size if $size > $max;
$s{$name} = [$size,$psize];
}
if ($max < 0) {
if ($USERWHERECLAUSE) {
add_ok 'T-EXCLUDE-USEROK';
}
else {
add_unknown 'T-EXCLUDE-DB';
}
next;
}
my $msg = '';
for (sort {$s{$b}[0] <=> $s{$a}[0] or $a cmp $b } keys %s) {
$msg .= "$_: $s{$_}[0] ($s{$_}[1]) ";
$db->{perf} .= " $_=$s{$_}[0]";
}
if (length $critical and $max >= $critical) {
add_critical $msg;
}
elsif (length $warning and $max >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_database_size
sub check_disk_space {
## Check the available disk space used by postgres
## Requires the executable "/bin/df"
## Must run as a superuser in the database (to examine 'data_directory' setting)
## Critical and warning are maximum size, or percentages
## Example: --critical="40 GB"
## NOTE: Needs to run on the same system (for now)
## XXX Allow custom ssh commands for remote df and the like
my ($warning, $critical) = validate_range
({
type => 'size or percent',
default_warning => '90%',
default_critical => '95%',
});
-x '/bin/df' or ndie qq{Could not find required executable /bin/df\n};
## Figure out where everything is.
$SQL = q{SELECT 'S', name, setting FROM pg_settings WHERE name = 'data_directory' }
. q{ OR name ='log_directory' }
. q{ UNION ALL }
. q{ SELECT 'T', spcname, spclocation FROM pg_tablespace WHERE spclocation <> ''};
my $info = run_command($SQL);
my %dir; ## 1 = normal 2 = been checked -1 = does not exist
my %seenfs;
for $db (@{$info->{db}}) {
my %i;
while ($db->{slurp} =~ /([ST])\s+\| (\w+)\s+\| (\S*)\s*/g) {
my ($st,$name,$val) = ($1,$2,$3);
$i{$st}{$name} = $val;
}
if (! exists $i{S}{data_directory}) {
add_unknown 'Could not determine data_directory: are you connecting as a superuser?';
next;
}
my ($datadir,$logdir) = ($i{S}{data_directory},$i{S}{log_directory}||'');
if (!exists $dir{$datadir}) {
if (! -d $datadir) {
add_unknown qq{could not find data directory "$datadir"};
$dir{$datadir} = -1;
next;
}
$dir{$datadir} = 1;
## Check if the WAL files are on a separate disk
my $xlog = "$datadir/pg_xlog";
if (-l $xlog) {
my $linkdir = readlink($xlog);
$dir{$linkdir} = 1 if ! exists $dir{$linkdir};
}
}
## Check log_directory: relative or absolute
if (length $logdir) {
if ($logdir =~ /^\w/) { ## relative, check only if symlinked
$logdir = "$datadir/$logdir";
if (-l $logdir) {
my $linkdir = readlink($logdir);
$dir{$linkdir} = 1 if ! exists $dir{$linkdir};
}
}
else { ## absolute, always check
if ($logdir ne $datadir and ! exists $dir{$logdir}) {
$dir{$logdir} = 1;
}
}
}
## Check all tablespaces
for my $tsname (keys %{$i{T}}) {
my $tsdir = $i{T}{$tsname};
$dir{$tsdir} = 1 if ! exists $dir{$tsdir};
}
my $gotone = 0;
for my $dir (keys %dir) {
next if $dir{$dir} != 1;
$dir{$dir} = 1;
$COM = "/bin/df -kP $dir 2>&1";
$res = qx{$COM};
if ($res !~ /^.+\n(\S+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\%\s+(\S+)/) {
ndie qq{Invalid result from command "$COM": $res\n};
}
my ($fs,$total,$used,$avail,$percent,$mount) = ($1,$2*1024,$3*1024,$4*1024,$5,$6);
## If we've already done this one, skip it
next if $seenfs{$fs}++;
next if skip_item($fs);
$gotone = 1;
## Rather than make another call with -h, do it ourselves
my $prettyused = pretty_size($used);
my $prettytotal = pretty_size($total);
my $msg = qq{FS $fs mounted on $mount is using $prettyused of $prettytotal ($percent%)};
$db->{perf} = "$fs=$used";
my $ok = 1;
if (length $critical) {
if (index($critical,'%')>=0) {
(my $critical2 = $critical) =~ s/\%//;
if ($percent >= $critical2) {
add_critical $msg;
$ok = 0;
}
}
elsif ($used >= $critical) {
add_critical $msg;
$ok = 0;
}
}
if (length $warning) {
if (index($warning,'%')>=0) {
(my $warning2 = $warning) =~ s/\%//;
if ($percent >= $warning2) {
add_warning $msg;
$ok = 0;
}
}
elsif ($used >= $warning) {
add_warning $msg;
$ok = 0;
}
}
if ($ok) {
add_ok $msg;
}
} ## end each dir
if (!$gotone) {
add_unknown 'T-EXCLUDE-FS';
}
}
return;
} ## end of check_disk_space
sub check_wal_files {
## Check on the number of WAL files in use
## Must run as a superuser
## Critical and warning are the number of files
## Example: --critical=40
my ($warning, $critical) = validate_range({type => 'integer', leastone => 1});
## Figure out where the pg_xlog directory is
$SQL = q{SELECT count(*) FROM pg_ls_dir('pg_xlog') WHERE pg_ls_dir ~ E'^[0-9A-F]{24}$'}; ## no critic
my $info = run_command($SQL);
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /(\d+)/) {
add_unknown qq{T-BAD-QUERY $db->{slurp}};
next;
}
my $numfiles = $1;
my $msg = qq{$numfiles};
if (length $critical and $numfiles > $critical) {
add_critical $msg;
}
elsif (length $warning and $numfiles > $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_wal_files
sub check_relation_size {
my $relkind = shift || 'relation';
## Check the size of one or more relations
## By default, checks all relations
## Can check specific one(s) with include
## Can ignore some with exclude
## Warning and critical are bytes
## Valid units: b, k, m, g, t, e
## All above may be written as plural or with a trailing 'g'
## Limit to a specific user (relation owner) with the includeuser option
## Exclude users with the excludeuser option
my ($warning, $critical) = validate_range({type => 'size'});
$VERBOSE >= 3 and warn "Warning and critical are now $warning and $critical\n";
$SQL = q{SELECT pg_relation_size(c.oid), pg_size_pretty(pg_relation_size(c.oid)), relkind, relname, nspname };
$SQL .= sprintf 'FROM pg_class c, pg_namespace n WHERE (relkind = %s) AND n.oid = c.relnamespace',
$relkind eq 'table' ? q{'r'} : $relkind eq 'index' ? q{'i'} : q{'r' OR relkind = 'i'};
if ($opt{perflimit}) {
$SQL .= " ORDER BY 1 DESC LIMIT $opt{perflimit}";
}
if ($USERWHERECLAUSE) {
$SQL =~ s/ WHERE/, pg_user u WHERE u.usesysid=c.relowner$USERWHERECLAUSE AND/;
}
my $info = run_command($SQL, {emptyok => 1});
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /\w/ and $USERWHERECLAUSE) {
add_ok 'T-EXCLUDE-USEROK';
next;
}
if ($db->{slurp} !~ /\d+\s+\|\s+\d+/) {
add_unknown "T-BAD-QUERY $db->{slurp}";
next;
}
my ($max,$pmax,$kmax,$nmax) = (-1,0,0,'?');
SLURP: while ($db->{slurp} =~ /(\d+) \| (\d+ \w+)\s+\| (\w)\s*\| (\S+)\s+\| (\S+)/gsm) {
my ($size,$psize,$kind,$name,$schema) = ($1,$2,$3,$4,$5);
next SLURP if skip_item($name);
$db->{perf} .= " $schema.$name=$size";
($max=$size, $pmax=$psize, $kmax=$kind, $nmax=$name) if $size > $max;
}
if ($max < 0) {
add_unknown 'T-EXCLUDE-REL';
next;
}
my $msg = sprintf qq{largest %s is %s"$nmax": $pmax},
$relkind, $relkind eq 'relation' ? ($kmax eq 'r' ? 'table ' : 'index ') : '';
if (length $critical and $max >= $critical) {
add_critical $msg;
}
elsif (length $warning and $max >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_relations_size
sub check_table_size {
return check_relation_size('table');
}
sub check_index_size {
return check_relation_size('index');
}
sub check_last_vacuum_analyze {
my $type = shift || 'vacuum';
my $auto = shift || 0;
## Check the last time things were vacuumed or analyzed
## NOTE: stats_row_level must be set to on in your database
## By default, reports on the oldest value in the database
## Can exclude and include tables
## Warning and critical are times, default to seconds
## Valid units: s[econd], m[inute], h[our], d[ay]
## All above may be written as plural as well (e.g. "2 hours")
## Limit to a specific user (relation owner) with the includeuser option
## Exclude users with the excludeuser option
## Example:
## --exclude=~pg_ --include=pg_class,pg_attribute
my ($warning, $critical) = validate_range
({
type => 'time',
default_warning => '1 day',
default_critical => '2 days',
});
my $criteria = $auto ?
qq{pg_stat_get_last_auto${type}_time(c.oid)}
: qq{GREATEST(pg_stat_get_last_${type}_time(c.oid), pg_stat_get_last_auto${type}_time(c.oid))};
## Do include/exclude earlier for large pg_classes?
$SQL = q{SELECT nspname, relname, CASE WHEN v IS NULL THEN -1 ELSE round(extract(epoch FROM now()-v)) END, }
.qq{ CASE WHEN v IS NULL THEN '?' ELSE TO_CHAR(v, '$SHOWTIME') END FROM (}
.qq{SELECT nspname, relname, $criteria AS v FROM pg_class c, pg_namespace n }
.q{WHERE relkind = 'r' AND n.oid = c.relnamespace ORDER BY 2) AS foo};
if ($opt{perflimit}) {
$SQL .= " ORDER BY 3 DESC LIMIT $opt{perflimit}";
}
if ($USERWHERECLAUSE) {
$SQL =~ s/ WHERE/, pg_user u WHERE u.usesysid=c.relowner$USERWHERECLAUSE AND/;
}
my $info = run_command($SQL, { regex => qr{\S+\s+\| \S+\s+\|}, emptyok => 1 } );
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /\w/ and $USERWHERECLAUSE) {
add_ok 'T-EXCLUDE-USEROK';
next;
}
my $maxtime = -2;
my $maxptime = '?';
my $maxrel = '?';
SLURP: while ($db->{slurp} =~ /(\S+)\s+\| (\S+)\s+\|\s+(\-?\d+) \| (.+)\s*$/gm) {
my ($schema,$name,$time,$ptime) = ($1,$2,$3,$4);
next SLURP if skip_item($name);
$db->{perf} .= " $schema.$name=$time" if $time >= 0;
if ($time > $maxtime) {
$maxtime = $time;
$maxrel = $name;
$maxptime = $ptime;
}
}
if ($maxtime == -2) {
add_unknown 'T-EXCLUDE-TABLES';
}
elsif ($maxtime == -1) {
add_unknown sprintf "No matching tables have ever been $type%s",
$type eq 'vacuum' ? 'ed' : 'd';
}
else {
my $msg = "$maxrel: $maxptime ($maxtime s)";
if ($critical and $maxtime >= $critical) {
add_critical $msg;
}
elsif ($warning and $maxtime >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
}
return;
} ## end of check_last_vacuum_analyze
sub check_last_vacuum {
my $auto = shift || '';
return check_last_vacuum_analyze('vacuum', $auto);
}
sub check_last_analyze {
my $auto = shift || '';
return check_last_vacuum_analyze('analyze', $auto);
}
sub check_listener {
## Check for a specific listener
## Critical and warning are simple strings, or regex if starts with a ~
## Example: --critical="~bucardo"
my ($warning, $critical) = validate_range({type => 'restringex'});
my $string = length $critical ? $critical : $warning;
my $regex = ($string =~ s/^~//) ? '~' : '=';
$SQL = "SELECT count(*) FROM pg_listener WHERE relname $regex '$string'";
my $info = run_command($SQL);
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /(\d+)/) {
add_unknown "T-BAD_QUERY $db->{slurp}";
next;
}
my $count = $1;
$db->{perf} .= " listening=$count";
my $msg = "listeners found: $count";
if ($count >= 1) {
add_ok $msg;
}
elsif ($critical) {
add_critical $msg;
}
else {
add_warning $msg;
}
}
return;
} ## end of check_listener
sub check_locks {
## Check the number of locks
## By default, checks all databases
## Can check specific databases with include
## Can ignore databases with exclude
## Warning and critical are either simple numbers, or more complex:
## Use locktype=number;locktype2=number
## The locktype can be "total", "waiting", or the name of a lock
## Lock names are case-insensitive, and do not need the "lock" at the end.
## Example: --warning=100 --critical="total=200;exclusive=20;waiting=5"
my ($warning, $critical) = validate_range
({
type => 'multival',
default_warning => 100,
default_critical => 150,
});
$SQL = q{SELECT granted, mode, datname FROM pg_locks l JOIN pg_database d ON (d.oid=l.database)};
my $info = run_command($SQL, { regex => qr[\s*\w+\s*\|\s*] });
for $db (@{$info->{db}}) {
my $gotone = 0;
my %lock = (total => 0);
my %dblock;
SLURP: while ($db->{slurp} =~ /([tf])\s*\|\s*(\w+)\s*\|\s*(\w+)\s+/gsm) {
my ($granted,$mode,$dbname) = ($1,lc $2,$3);
next SLURP if skip_item($dbname);
$gotone = 1;
$lock{total}++;
$mode =~ s{lock$}{};
$lock{$mode}++;
$lock{waiting}++ if $granted ne 't';
$lock{$dbname}++; ## We assume nobody names their db 'rowexclusivelock'
$dblock{$dbname}++;
}
for (sort keys %dblock) {
$db->{perf} .= " $_=$dblock{$_}";
}
if (!$gotone) {
add_unknown 'T-EXCLUDE-DB';
}
## If not specific errors, just use the total
my $ok = 1;
if (ref $critical) {
for my $type (keys %lock) {
next if ! exists $critical->{$type};
if ($lock{$type} >= $critical->{$type}) {
add_critical qq{total "$type" locks: $lock{$type}};
$ok = 0;
}
}
}
elsif (length $critical and $lock{total} >= $critical) {
add_critical qq{total locks: $lock{total}};
$ok = 0;
}
if (ref $warning) {
for my $type (keys %lock) {
next if ! exists $warning->{$type};
if ($lock{$type} >= $warning->{$type}) {
add_warning qq{total "$type" locks: $lock{$type}};
$ok = 0;
}
}
}
elsif (length $warning and $lock{total} >= $warning) {
add_warning qq{total locks: $lock{total}};
$ok = 0;
}
if ($ok) {
my %show;
if (!keys %critical and !keys %warning) {
$show{total} = 1;
}
for my $type (keys %critical) {
$show{$type} = 1;
}
for my $type (keys %warning) {
$show{$type} = 1;
}
my $msg = '';
for (sort keys %show) {
$msg .= sprintf "$_=%d ", $lock{$_} || 0;
}
add_ok $msg;
}
}
return;
} ## end of check_locks
sub check_logfile {
## Make sure the logfile is getting written to
## Especially useful for syslog redirectors
## Should be run on the system housing the logs
## Optional argument "logfile" tells where the logfile is
## Allows for some conversion characters.
## Example: --logfile="/syslog/%Y-m%-d%/H%/postgres.log"
## Critical and warning are not used: it's either ok or critical.
my $critwarn = $opt{warning} ? 0 : 1;
$SQL = q{SELECT CASE WHEN length(setting)<1 THEN '?' ELSE setting END FROM pg_settings WHERE name };
$SQL .= q{IN ('log_destination','log_directory','log_filename','redirect_stderr','syslog_facility') ORDER BY name};
my $logfilere = qr{^[\w_\s\/%\-\.]+$};
if (exists $opt{logfile} and $opt{logfile} !~ $logfilere) {
ndie qq{Invalid logfile option\n};
}
my $info = run_command($SQL);
$VERBOSE >= 3 and warn Dumper $info;
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /^\s*(\w+)\n\s*(.+?)\n\s*(.+?)\n\s*(\w*)\n\s*(\w*)/sm) {
add_unknown "T-BAD-QUERY $db->{slurp}";
next;
}
my ($dest,$dir,$file,$redirect,$facility) = ($1,$2,$3,$4,$5||'?');
$VERBOSE >=3 and warn "Dest is $dest, dir is $dir, file is $file, facility is $facility\n";
## Figure out what we think the log file will be
my $logfile ='';
if (exists $opt{logfile} and $opt{logfile} =~ /\w/) {
$logfile = $opt{logfile};
} else {
if ($dest eq 'syslog') {
## We'll make a best effort to figure out where it is. Using the --logfile option is preferred.
$logfile = '/var/log/messages';
if (open my $cfh, '<', '/etc/syslog.conf') {
while (<$cfh>) {
if (/\b$facility\.(?!none).+?([\w\/]+)$/i) {
$logfile = $1;
}
}
}
if (!$logfile or ! -e $logfile) {
ndie "Database is using syslog, please specify path with --logfile option (fac=$facility)\n";
}
} elsif ($dest eq 'stderr') {
if ($redirect ne 'yes') {
ndie qq{Logfile output has been redirected to stderr: please provide a filename\n};
}
}
}
## We now have a logfile (or a template)..parse it into pieces.
## We need at least hour, day, month, year
my @t = localtime $^T;
my ($H,$d,$m,$Y) = (sprintf ('%02d',$t[2]),sprintf('%02d',$t[3]),sprintf('%02d',$t[4]+1),$t[5]+1900);
if ($logfile !~ $logfilere) {
ndie qq{Invalid logfile "$logfile"\n};
}
$logfile =~ s/%%/~~/g;
$logfile =~ s/%Y/$Y/g;
$logfile =~ s/%m/$m/g;
$logfile =~ s/%d/$d/g;
$logfile =~ s/%H/$H/g;
$VERBOSE >= 3 and warn "Final logfile: $logfile\n";
if (! -e $logfile) {
if ($critwarn) {
add_unknown qq{logfile "$logfile" does not exist!};
}
else {
add_warning qq{logfile "$logfile" does not exist!};
}
next;
}
my $logfh;
unless (open $logfh, '<', $logfile) {
add_unknown qq{logfile "$logfile" failed to open: $!\n};
next;
}
seek($logfh, 0, 2) or ndie qq{Seek on $logfh failed: $!\n};
## Throw a custom error string
my $smallsearch = sprintf 'Random=%s', int rand(999999999999);
my $funky = sprintf "$ME this_statement_will_fail DB=$db->{dbname} PID=$$ Time=%s $smallsearch",
scalar localtime;
## Cause an error on just this target
delete $db->{ok}; delete $db->{slurp}; delete $db->{totaltime};
my $badinfo = run_command("SELECT $funky", {failok => 1, target => $db} );
my $MAXSLEEPTIME = 3;
my $SLEEP = 0.5;
my $found = 0;
LOGWAIT: {
sleep $SLEEP;
seek $logfh, 0, 1 or ndie qq{Seek on $logfh failed: $!\n};
while (<$logfh>) {
if (/$smallsearch/) { ## Some logs break things up, so we don't use funky
$found = 1;
last LOGWAIT;
}
}
$MAXSLEEPTIME -= $SLEEP;
redo if $MAXSLEEPTIME > 0;
if ($critwarn) {
add_critical qq{fails logging to: $logfile};
}
else {
add_warning qq{fails logging to: $logfile};
}
}
close $logfh or ndie qq{Could not close $logfh: $!\n};
if ($found == 1) {
add_ok qq{logs to: $logfile};
}
}
return;
} ## end of check_logfile
sub check_query_runtime {
## Make sure a known query runs at least as fast as we think it should
## Warning and critical are time limits, defaulting to seconds
## Valid units: s[econd], m[inute], h[our], d[ay]
## Does a "EXPLAIN ANALYZE SELECT COUNT(1) FROM xyz"
## where xyz is given by the option --queryname
## This could also be a table or a function, or course, but must be a
## single word. If a function, it must be empty (with "()")
## Examples:
## --warning="100s" --critical="120s" --queryname="speedtest1"
## --warning="5min" --critical="15min" --queryname="speedtest()"
my ($warning, $critical) = validate_range({type => 'time'});
my $queryname = $opt{queryname} || '';
if ($queryname !~ /^[\w\.]+(?:\(\))?$/) {
ndie q{Invalid queryname option: must be a simple view name};
}
$SQL = "EXPLAIN ANALYZE SELECT COUNT(1) FROM $queryname";
my $info = run_command($SQL);
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /Total runtime: (\d+\.\d+) ms\s*$/s) {
add_unknown "T-BAD-QUERY $db->{slurp}";
next;
}
my $totalseconds = $1 / 1000.0;
$db->{perf} = " qtime=$totalseconds";
my $msg = qq{query runtime: $totalseconds seconds};
if (length $critical and $totalseconds >= $critical) {
add_critical $msg;
}
elsif (length $warning and $totalseconds >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_query_runtime
sub check_query_time {
## Check the length of running queries
## It makes no sense to run this more than once on the same cluster
## Warning and critical are time limits - defaults to seconds
## Valid units: s[econd], m[inute], h[our], d[ay]
## All above may be written as plural as well (e.g. "2 hours")
## Can also ignore databases with exclude and limit with include
## Limit to a specific user with the includeuser option
## Exclude users with the excludeuser option
my ($warning, $critical) = validate_range
({
type => 'time',
default_warning => '2 minutes',
default_critical => '5 minutes',
});
$SQL = q{SELECT datname, max(COALESCE(ROUND(EXTRACT(epoch FROM now()-query_start)),0)) }.
qq{FROM pg_stat_activity WHERE current_query <> '<IDLE>'$USERWHERECLAUSE GROUP BY 1};
my $info = run_command($SQL, { regex => qr{\s*.+?\s+\|\s+\d+}, emptyok => 1 } );
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /\w/ and $USERWHERECLAUSE) {
add_ok 'T-EXCLUDE-USEROK';
next;
}
my $max = -1;
SLURP: while ($db->{slurp} =~ /(.+?)\s+\|\s+(\d+)\s*/gsm) {
my ($dbname,$current) = ($1,$2);
next SLURP if skip_item($dbname);
$max = $current if $current > $max;
}
$db->{perf} .= " maxtime:$max";
if ($max < 0) {
add_unknown 'T-EXCLUDE-DB';
next;
}
my $msg = qq{longest query: ${max}s};
if (length $critical and $max >= $critical) {
add_critical $msg;
}
elsif (length $warning and $max >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_query_time
sub check_txn_time {
## Check the length of running transactions
## It makes no sense to run this more than once on the same cluster
## Warning and critical are time limits - defaults to seconds
## Valid units: s[econd], m[inute], h[our], d[ay]
## All above may be written as plural as well (e.g. "2 hours")
## Can also ignore databases with exclude and limit with include
## Limit to a specific user with the includeuser option
## Exclude users with the excludeuser option
my ($warning, $critical) = validate_range
({
type => 'time',
});
$SQL = q{SELECT datname, max(COALESCE(ROUND(EXTRACT(epoch FROM now()-xact_start)),0)) }.
qq{FROM pg_stat_activity WHERE xact_start IS NOT NULL$USERWHERECLAUSE GROUP BY 1};
my $info = run_command($SQL, { regex => qr[\s*.+?\s+\|\s+\d+], emptyok => 1 } );
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /\w/ and $USERWHERECLAUSE) {
add_ok 'T-EXCLUDE-USEROK';
next;
}
my $max = -1;
SLURP: while ($db->{slurp} =~ /(.+?)\s+\|\s+(\d+)\s*/gsm) {
my ($dbname,$current) = ($1,$2);
next SLURP if skip_item($dbname);
$max = $current if $current > $max;
}
$db->{perf} .= " maxtime:$max";
if ($max < 0) {
add_unknown 'T-EXCLUDE-DB';
next;
}
my $msg = qq{longest txn: ${max}s};
if (length $critical and $max >= $critical) {
add_critical $msg;
}
elsif (length $warning and $max >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_txn_time
sub check_txn_idle {
## Check the length of "idle in transaction" connections
## It makes no sense to run this more than once on the same cluster
## Warning and critical are time limits - defaults to seconds
## Valid units: s[econd], m[inute], h[our], d[ay]
## All above may be written as plural as well (e.g. "2 hours")
## Can also ignore databases with exclude and limit with include
## Limit to a specific user with the includeuser option
## Exclude users with the excludeuser option
my ($warning, $critical) = validate_range
({
type => 'time',
});
$SQL = q{SELECT datname, max(COALESCE(ROUND(EXTRACT(epoch FROM now()-query_start)),0)) }.
qq{FROM pg_stat_activity WHERE current_query = '<IDLE> in transaction'$USERWHERECLAUSE GROUP BY 1};
my $info = run_command($SQL, { regex => qr[\s*.+?\s+\|\s+\d+], emptyok => 1 } );
for $db (@{$info->{db}}) {
my $max = -1;
if ($db->{slurp} !~ /\w/ and $USERWHERECLAUSE) {
add_ok 'T-EXCLUDE-USEROK';
next;
}
if ($db->{slurp} =~ /^\s*$/o) {
add_ok 'no idle in transaction';
next;
}
SLURP: while ($db->{slurp} =~ /(.+?)\s+\|\s+(\d+)\s*/gsm) {
my ($dbname,$current) = ($1,$2);
next SLURP if skip_item($dbname);
$max = $current if $current > $max;
}
$db->{perf} .= " maxtime:$max";
if ($max < 0) {
add_unknown 'T-EXCLUDE-DB';
next;
}
my $msg = qq{longest idle in txn: ${max}s};
if (length $critical and $max >= $critical) {
add_critical $msg;
}
elsif (length $warning and $max >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_txn_idle
sub check_settings_checksum {
## Verify the checksum of all settings
## Not that this will vary from user to user due to ALTER USER
## and because superusers see additional settings
## One of warning or critical must be given (but not both)
## It should run one time to find out the expected checksum
## You can use --critical="0" to find out the checksum
## You can include or exclude settings as well
## Example:
## check_settings_checksum --critical="4e7ba68eb88915d3d1a36b2009da4acd"
my ($warning, $critical) = validate_range({type => 'checksum', onlyone => 1});
eval {
require Digest::MD5;
};
if ($@) {
ndie qq{Sorry, you must install the Perl module Digest::MD5 first\n};
}
$SQL = 'SELECT name, setting FROM pg_settings ORDER BY name';
my $info = run_command($SQL, { regex => qr[client_encoding] });
for $db (@{$info->{db}}) {
(my $string = $db->{slurp}) =~ s/\s+$/\n/;
my $newstring = '';
SLURP: for my $line (split /\n/ => $string) {
ndie q{Invalid pg_setting line} unless $line =~ /^\s*(\w+)/;
my $name = $1;
next SLURP if skip_item($name);
$newstring .= "$line\n";
}
if (! length $newstring) {
add_unknown 'T-EXCLUDE-SET';
}
my $checksum = Digest::MD5::md5_hex($newstring);
my $msg = "checksum: $checksum";
if ($critical and $critical ne $checksum) {
add_critical $msg;
}
elsif ($warning and $warning ne $checksum) {
add_warning $msg;
}
elsif (!$critical and !$warning) {
add_unknown $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_settings_checksum
sub check_timesync {
## Compare local time to the database time
## Warning and critical are given in number of seconds difference
my ($warning,$critical) = validate_range
({
type => 'seconds',
default_warning => 2,
default_critical => 5,
});
$SQL = q{SELECT round(extract(epoch FROM now())), TO_CHAR(now(),'YYYY-MM-DD HH24:MI:SS')};
my $info = run_command($SQL);
my $localepoch = time;
my @l = localtime;
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /(\d+) \| (.+)/) {
add_unknown "T-BAD-QUERY $db->{slurp}";
next;
}
my ($pgepoch,$pgpretty) = ($1,$2);
my $diff = abs($pgepoch - $localepoch);
$db->{perf} = " diff:$diff";
my $localpretty = sprintf '%d-%02d-%02d %02d:%02d:%02d', $l[5]+1900, $l[4], $l[3],$l[2],$l[1],$l[0];
my $msg = qq{timediff=$diff DB=$pgpretty Local=$localpretty};
if (length $critical and $diff >= $critical) {
add_critical $msg;
}
elsif (length $warning and $diff >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
return;
} ## end of check_timesync
sub check_txn_wraparound {
## Check how close to transaction wraparound we are on all databases
## Warning and critical are the number of transactions left
## See: http://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND
## It makes no sense to run this more than once on the same cluster
my ($warning, $critical) = validate_range
({
type => 'positive integer',
default_warning => 1_300_000_000,
default_critical => 1_400_000_000,
});
$SQL = q{SELECT datname, age(datfrozenxid) FROM pg_database WHERE datallowconn is true};
my $info = run_command($SQL, { regex => qr[\w+\s+\|\s+\d+] } );
for $db (@{$info->{db}}) {
while ($db->{slurp} =~ /(\S+)\s+\|\s+(\d+)/gsm) {
my ($dbname,$dbtxns) = ($1,$2);
my $msg = qq{$dbname: $dbtxns};
$db->{perf} .= " $dbname=$dbtxns";
$VERBOSE >= 3 and warn $msg;
if (length $critical and $dbtxns >= $critical) {
add_critical $msg;
}
elsif (length $warning and $dbtxns >= $warning) {
add_warning $msg;
}
else {
add_ok $msg;
}
}
}
return;
} ## end of check_txn_wraparound
sub check_version {
## Compare version with what we think it should be
## Warning and critical are the major and minor (e.g. 8.3)
## or the major, minor, and revision (e.g. 8.2.4 or even 8.3beta4)
my ($warning, $critical) = validate_range({type => 'version'});
my ($warnfull, $critfull) = (($warning =~ /^\d+\.\d+$/ ? 0 : 1),($critical =~ /^\d+\.\d+$/ ? 0 : 1));
my $info = run_command('SELECT version()');
for $db (@{$info->{db}}) {
if ($db->{slurp} !~ /PostgreSQL ((\d+\.\d+)(\w+|\.\d+))/o) {
add_unknown "T-BAD-QUERY $db->{slurp}";
next;
}
my ($full,$version,$revision) = ($1,$2,$3||'?');
$revision =~ s/^\.//;
my $ok = 1;
if (length $critical) {
if (($critfull and $critical ne $full)
or (!$critfull and $critical ne $version)) {
add_critical qq{version $full, but expected $critical};
$ok = 0;
}
}
elsif (length $warning) {
if (($warnfull and $warning ne $full)
or (!$warnfull and $warning ne $version)) {
add_warning qq{version $full, but expected $warning};
$ok = 0;
}
}
if ($ok) {
add_ok "version $full";
}
}
return;
} ## end of check_version
__END__
=pod
=head1 NAME
check_postgres.pl - Postgres monitoring script for Nagios
=head1 VERSION
This documents describes B<check_postgres.pl> version 1.5.2
=head1 SYNOPSIS
## Create all symlinks
check_postgres.pl --action=build_symlinks
## Check connection to Postgres database 'pluto':
check_postgres.pl --action=connection --db=pluto
## Same things, but using the symlink
check_postgres_connection --db=pluto
## Warn if > 100 locks, critical if > 200, or > 20 exclusive
check_postgres_locks --warning=100 --critical="total=200;exclusive=20"
## There are many other actions and options, please keep reading.
=head1 WEBSITE
The latest news and documentation can always be found at:
http://bucardo.org/check_postgres/
=head1 DESCRIPTION
check_postgres.pl is a Perl script that runs many different tests against
one or more Postgres databases. It uses the psql program to gather the
information, and returns one of four exit codes used by Nagios, as well
as a short description of the results. The exit codes are:
=over 2
=item 0 (OK)
=item 1 (WARNING)
=item 2 (CRITICAL)
=item 3 (UNKNOWN)
=back
=head1 DATABASE CONNECTION OPTIONS
Almost all actions accept a common set of options, most dealing with connecting to the databases.
=over 4
=item B<-H NAME> or B<--host=NAME>
Connect to the host indicated by NAME. Can be a comma-separated list of names. Multiple host arguments
are allowed. If no host is given, defaults to a local Unix socket.
=item B<-p PORT> or B<--port=PORT>
Connects using the specified PORT number. Can be a comma-separated list of port numbers, and multiple
port arguments are allowed. If no port number is given, we default to port 5432.
=item B<-db NAME> or B<--dbname=NAME>
Specifies which database to connect to. Can be a comma-separated list of names, and multiple dbname
arguments are allowed. If no dbname option is provided, defaults to 'postgres' if the psql
version is version 8 or greater, and 'template1' otherwise.
=item B<-u USERNAME> or B<--dbuser=USERNAME>
The name of the database user to connect as. Can be a comma-separated list of usernames, and multiple
dbuser arguments are allowed. If this is not provided, defaults to 'postgres'.
=item B<--dbpass=PASSWORD>
Provides the password to connect to the database with. Use of this option is highly discouraged.
Instead, one should use a .pgpass file.
=back
Connection options can be grouped: --host=a,b --host=c --port=1234 --port=3344
would connect to a-1234, b-1234, and c-3344. Note that once set, an option
carries over until it is changed again.
Examples:
--host=a,b --port=5433 --db=c
Connects twice to port 5433, using database c, to hosts a and b
a-5433-c b-5433-c
--host=a,b --port=5433 --db=c,d
Connects four times: a-5433-c a-5433-d b-5433-c b-5433-d
--host=a,b --host=foo --port=1234 --port=5433 --db=e,f
Connects six times: a-1234-e a-1234-f b-1234-e b-1234-f foo-5433-e foo-5433-f
--host=a,b --host=x --port=5432,5433 --dbuser=alice --dbuser=bob -db=baz
Connects three times: a-5432-alice-baz b-5433-alice-baz x-5433-bob-baz
=head1 OTHER OPTIONS
Other common options include:
=over 4
=item B<--output=VAL>
Determines the format of the output, for use in various programs. The default is 'nagios'. No
other systems are supported yet.
=item B<PSQL=PATH>
Tells the script where to find the psql program. Useful if you have more than
one version of the psql executable on your system, or if there is no psql program
in your path. Note that this option is in all uppercase. By default, this option
is I<not allowed>. To enable it, you must change the C<$NO_PSQL_OPTION> near the
top of the script to 0. Avoid using this option if you can, and instead hard-code
your psql location into the C<$PSQL> variable, also near the top of the script.
=item B<-t VAL> or B<--timeout=VAL>
Sets the timeout in seconds after which the script will abort whatever it is doing
and return an UNKNOWN status. The timeout is per Postgres cluster, not for the entire
script. The default value is 10; the units are always in seconds.
=item B<-h> or B<--help>
Displays a help screen with a summary of all actions and options.
=item B<-V> or B<--version>
Shows the current version.
=item B<-v> or B<--verbose>
Set the verbosity level. Can call more than once to boost the level. Setting it to three
or higher (in other words, issuing C<-v -v -v>) turns on debugging information for this
program which is sent to stderr.
=item B<--test>
Enables test mode. See the L</"TEST MODE"> section below.
=item B<--showperf=VAL>
Determines if we output performance data in standard Nagios format
(at end of string, after a pipe symbol, using name=value).
VAL should be 0 or 1. The default is 1.
=item B<--perflimit=i>
Sets a limit s to how many items of interest are reported back when using the
B<showperf> option. This only has an effect for actions that return a large
number of items, such as B<table_size>. The default is 0, or no limit. Be
careful when using this with the B<--include> or B<--exclude> options, as
those restrictions are done I<after> the query has been run, and thus your
limit may not include the items you want.
=item B<--showtime=VAL>
Determines if the time taken to run each query is shown in the output. VAL
should be 0 or 1. The default is 1. No effect unless B<showperf> is on.
=item B<--action=NAME>
States what action we are running. Required unless using a symlinked file,
in which case the name of the file is used to figure out the action.
=back
=head1 ACTIONS
The script runs one or more actions. This can either be done with the --action
flag, or by using a symlink to the main file that contains the name of the action
inside of it. For example, to run the action "timesync", you may either issue:
check_postgres.pl --action=timesync
or use a program named:
check_postgres_timesync
All the symlinks are created for you in the current directory
if use the action "build_symlinks":
perl check_postgres.pl --action="build_symlinks"
If the file name already exists, it will not be overwritten. If the file exists
and is a symlink, you can force it to overwrite by using "build_symlinks_force"
Most actions take a B<--warning> and an B<--critical> option, indicating at what
point we change from OK to WARNING, and what point we go to CRITICAL. Note that
because criticals are always checked first, setting the warning equal to the
critical is an effective way to turn warnings off and always give a critical.
The current supported actions are:
=over 4
=item B<backends> (symlink: C<check_postgres_backends>)
Checks the current number of connections for one or more databases, and optionally
compares it to the maximum allowed, which is determined by the
Postgres configuration variable B<max_connections>. The B<--warning> and
B<--critical> options can take one of three forms. First, a simple number can be
given, which represents the number of connections at which the alert will be
given. This choice does not use the B<max_connections> setting. Second, the
percentage of available connections can be given. Third, a negative number can
be given which represents the number of connections left until B<max_connections>
is reached. The default values for B<--warning> and B<--critical> are '90%' and '95%'.
You can also filter the databases by use of the
B<--include> and B<--exclude> options. See the L</"BASIC FILTERING"> section
for more details.
Example 1: Give a warning when the number of connections on host quirm reaches 120, and a critical if it reaches 140.
check_postgres_backends --host=quirm --warning=120 --critical=150
Example 2: Give a critical when we reach 75% of our max_connections setting on hosts lancre or lancre2.
check_postgres_backends --warning='75%' --critical='75%' --host=lancre,lancre2
Example 3: Give a warning when there are only 10 more connection slots left on host plasmid, and a critical
when we have only 5 left.
check_postgres_backends --warning=-10 --critical=-5 --host=plasmid
Example 4: Check all databases except those with "test" in their name, but allow ones that are named "pg_greatest". Connect as port 5432 on the first two hosts, and as port 5433 on the third one. We want to always throw a critical when we reach 30 or more connections.
check_postgres_backends --dbhost=hong,kong --dbhost=fooey --dbport=5432 --dbport=5433 --warning=30 --critical=30 --exclude="~test" --include="pg_greatest,~prod"
=item B<bloat> (symlink: C<check_postgres_bloat>)
Checks the amount of bloat in tables and indexes. (Bloat is generally the amount
of dead unused space taken up in a table or index. This space is usually reclaimed
by use of the VACUUM command.) This action requires that stats collection be
enabled on the target databases, and requires that ANALYZE is run frequently.
The B<--include> and B<--exclude> options can be used to filter out which tables
to look at. See the L</"BASIC FILTERING"> section for more details.
The B<--warning> and B<--critical> options must be specified as sizes.
Valid units are bytes, kilobytes, megabytes, gigabytes, terabytes, and exabytes.
You can abbreviate all of those with the first letter. Items without units are
assumed to be 'bytes'. The default values are '1 GB' and '5 GB'. The value
represents the number of "wasted bytes", or the difference between what is actually
used by the table and index, and what we compute that it should be.
Note that this action has two hard-coded values to avoid false alarms on
smaller relations. Tables must have at least 10 pages, and indexes at least 15,
before they can be considered by this test. If you really want to adjust these
values, you can look for the variables B<$MINPAGES> and B<$MINIPAGES> at the top of the
C<check_bloat> subroutine.
Please note that the values computed by this action are not precise, and
should be used as a guideline only. Great effort was made to estimate the
correct size of a table, but in the end it is only an estimate. The correct
index size is even more of a guess than the correct table size, but both
should give a rough idea of how bloated things are.
Example 1: Warn if any table on port 5432 is over 100 MB bloated, and critical if over 200 MB
check_postgres_bloat --port=5432 --warning='100 M', --critical='200 M'
Example 2: Give a critical if table 'orders' on host 'sami' has more than 10 megs of bloat
check_postgres_bloat --host=sami --include=orders --critical='10 MB'
=item B<connection> (symlink: check_postgres_connection)
Simply connects, issues a 'SELECT version()', and leaves.
Takes no B<--warning> or B<--critical> options.
=item B<database_size> (symlink: C<check_postgres_database_size>)
Checks the size of all databases and complains when they are too big.
There is no need to run this command more than once per database cluster.
Databases can be filtered with
the B<--include> and B<--exclude> options. See the L</"BASIC FILTERING"> section
for more details.
They can also be filtered by the owner of the database with the
B<--includeuser> and B<--excludeuser> options.
See the L</"USER NAME FILTERING"> section for more details.
The warning and critical options can be specified as bytes, kilobytes, megabytes,
gigabytes, terabytes, or exabytes. Each may be abbreviated to the first letter as well.
If no unit is given, the units are assumed to be bytes. There are not defaults for this
action: the warning and critical must be specified. The warning value cannot be greater
than the critical value. The output returns all databases sorted by size largest first,
showing both raw bytes and a "pretty" version of the size.
Example 1: Warn if any database on host flagg is over 1 TB in size, and critical if over 1.1 TB.
check_postgres_database_size --host=flagg --warning='1 TB' --critical='1.1 t'
Example 2: Give a critical if the database template1 on port 5432 is over 10 MB.
check_postgres_database_size --port=5432 --include=template1 --warning='10MB' --critical='10MB'
Example 3: Give a warning if any database on host 'tardis' owned by the user 'tom' is over 5 GB
check_postgres_database_size --host=tardis --includeuser=tom --warning='5 GB' --critical='10 GB'
=item B<disk_space> (symlink: C<check_postgres_disk_space>)
Checks on the available physical disk space used by Postgres. This action requires
that you have the executable "/bin/df" available to report on disk sizes, and it
also needs to be run as a superuser, so it can examine the B<data_directory>
setting inside of Postgres. The B<--warning> and B<--critical> options are
given in either sizes or percentages. If using sizes, the standard unit types
are allowed: bytes, kilobytes, gigabytes, megabytes, gigabytes, terabytes, or
exabytes. Each may be abbreviated to the first letter only; no units at all
indicates 'bytes'. The default values are '90%' and '95%'.
This command checks the following things to determine all of the different
physical disks being used by Postgres.
B<data_directory> - The disk that the main data directory is on.
B<log directory> - The disk that the log files are on.
B<WAL file directory> - The disk that the write-ahead logs are on (e.g. symlinked pg_xlog)
B<tablespaces> - Each tablespace that is on a separate disk.
The output shows the total size used and available on each disk, as well as
the percentage, ordered by highest to lowest percentage used. Each item above
maps to a file system: these can be included or excluded. See the
L</"BASIC FILTERING"> section for more details.
Example 1: Make sure that no file system is over 90% for the database on port 5432.
check_postgres_disk_space --port=5432 --warning='90%' --critical="90%'
Example 2: Check that all file systems starting with /dev/sda are smaller than 10 GB and 11 GB (warning and critical)
check_postgres_disk_space --port=5432 --warning='10 GB' --critical='11 GB' --include="~^/dev/sda"
=item B<index_size> (symlink: C<check_postgres_index_size>)
=item B<table_size> (symlink: C<check_postgres_table_size>)
=item B<relation_size> (symlink: C<check_postgres_relation_size>)
The actions B<table_size> and B<index_size> are simply variations of the
B<relation_size> action, which checks for a relation that has grown too big.
Relations (in other words, tables and indexes) can be filtered with the
B<--include> and B<--exclude> options. See the L</"BASIC FILTERING"> section
for more details. Relations can also be filtered by the user that owns them,
by using the B<--includeuser> and B<--excludeuser> options.
See the L</"USER NAME FILTERING"> section for more details.
The values for the B<--warning> and B<--critical> options are file sizes, and
may have units of bytes, kilobytes, megabytes, gigabytes, terabytes, or exabytes.
Each can be abbreviated to the first letter. If no units are given, bytes are
assumed. There are no default values: both the warning and the critical option
must be given. The return text shows the size of the largest relation found.
If the B<--showperf> option is enabled, I<all> of the relations with their sizes
will be given. To prevent this, it is recommended that you set the
B<--perflimit> option, which will cause the query to do a
C<ORDER BY size DESC LIMIT (perflimit)>.
Example 1: Give a critical if any table is larger than 600MB on host burrick.
check_postgres_table_size --critical='600 MB' --warning='600 MB' --host=burrick
Example 2: Warn if the table products is over 4 GB in size, and give a critical at 4.5 GB.
check_postgres_table_size --host=burrick --warning='4 GB' --critical='4.5 GB' --include=products
Example 3: Warn if any index not owned by postgres goes over 500 MB.
check_postgres_index_size --port=5432 --excludeuser=postgres -w 500MB -c 600MB
=item B<last_analyze> (symlink: C<check_postgres_last_analyze>)
=item B<last_vacuum> (symlink: C<check_postgres_last_vacuum>)
Checks how long it has been since vacuum (or analyze) was last run on each
table in one or more databases. Use of these actions requires that the Postgres
configuration variable B<stats_rows_level> is enabled, and that the target
database is version 8.2 or higher. Tables can be filtered with the
B<--include> and B<--exclude> options. See the L</"BASIC FILTERING"> section
for more details.
Tables can also be filtered by their owner by use of the
B<--includeuser> and B<--excludeuser> options.
See the L</"USER NAME FILTERING"> section for more details.
The units for B<--warning> and B<--critical> are specified as times.
Valid units are seconds, minutes, hours, and days; all can be abbreviated
to the first letter. If no units are given, 'seconds' are assumed. The
default values are '1 day' and '2 days'. Please note that there are cases
in which this field does not get automatically populated. If certain tables
are giving you problems, make sure that they have dead rows to vacuum,
or just exclude them from the test.
Example 1: Warn if any table has not been vacuumed in 3 days, and give a
critical at a week, for host wormwood
check_last_vacuum --host=wormwood --warning='3d' --critical='7d'
Example 2: Same as above, but skip tables belonging to the users 'eve' or 'mallory'
check_last_vacuum --host=wormwood --warning='3d' --critical='7d' --excludeusers=eve,mallory
=item B<listener> (symlink: C<check_postgres_listener>)
Confirm that someone is listening for one or more specific strings. Only one of warning or critical is needed. The format
is a simple string representing the LISTEN target, or a tilde character followed by a string for a regular expression
check.
Example 1: Give a warning if nobody is listening for the string bucardo_mcp_ping on ports 5555 and 5556
check_postgres_listener --port=5555,5556 --warning=bucardo_mcp_ping
Example 2: Give a critical if there are no active LISTEN requests matching 'grimm' on database oskar
check_postgres_listener --db oskar --critical=~grimm
=item B<locks> (symlink: C<check_postgres_locks>)
Check the total number of locks on one or more databases. There is no
need to run this more than once per database cluster. Databases can be filtered
with the B<--include> and B<--exclude> options. See the L</"BASIC FILTERING"> section
for more details.
The B<--warning> and B<--critical> options can be specified as simple numbers,
which represent the total number of locks, or they can be broken down by type of lock.
Valid lock names are C<'total'>, C<'waiting'>, or the name of a lock type used by Postgres.
These names are case-insensitive and do not need the "lock" part on the end,
so B<exclusive> will match 'ExclusiveLock'. The format is name=number, with different
items separated by semicolons.
Example 1: Warn if the number of locks is 100 or more, and critical if 200 or more, on host garrett
check_postgres_locks --host=garrett --warning=100 --critical=200
Example 2: On the host artemus, warn if 200 or more locks exist, and give a critical if over 250 total locks exist, or if over 20 exclusive locks exist, or if over 5 connections are waiting for a lock.
check_postgres_locks --host=artemus --warning=200 --critical="total=250;waiting=5;exclusive=20"
=item B<logfile> (symlink: C<check_postgres_logfile>)
Ensures that the logfile is in the expected location and is being logged to.
This action issues a command that throws an error on each database it is
checking, and ensures that the message shows up in the logs. It scans the
various log_* settings inside of Postgres to figure out where the logs should be.
If you are using syslog, it does a rough (but not foolproof) scan of
F</etc/syslog.conf>. Alternatively, you can provide the name of the logfile
with the B<--logfile> option. This is especially useful if the logs have a
custom rotation scheme driven be an external program. The B<--logfile> option
supports the following escape characters: C<%Y %m %d %H>, which represent
the current year, month, date, and hour respectively. An error is always
reported as critical unless the warning option has been passed in as a non-zero
value. Other than that specific usage, the C<--warning> and C<--critical>
options should I<not> be used.
Example 1: On port 5432, ensure the logfile is being written to the file /home/greg/pg8.2.log
check_postgres_logfile --port=5432 --logfile=/home/greg/pg8.2.log
Example 2: Same as above, but raise a warning, not a critical
check_postgres_logfile --port=5432 --logfile=/home/greg/pg8.2.log -w 1
=item B<query_runtime> (symlink: C<check_postgres_query_runtime>)
Checks how long a specific query takes to run, by executing a "EXPLAIN ANALYZE"
against it. The B<--warning> and B<--critical> options are the maximum amount of
time the query should take. Valid units are seconds, minutes, and hours; any can be
abbreviated to the first letter. If no units are given, 'seconds' are assumed.
Both the warning and the critical option must be given. The name of the view or
function to be run must be passed in to the B<--queryname> option. It must consist
of a single word (or schema.word), with optional parens at the end.
Example 1: Give a critical if the function named "speedtest" fails to run in 10 seconds or less.
check_postgres_query_runtime --queryname='speedtest()' --critical=10 --warning=10
=item B<query_time> (symlink: C<check_postgres_query_time>)
Checks the length of running queries on one or more databases. There is
no need to run this more than once on the same database cluster.
Databases can be filtered
by using the B<--include> and B<--exclude> options. See the L</"BASIC FILTERING">
section for more details. You can also filter on the user running the
query with the B<--includeuser> and B<--excludeuser> options.
See the L</"USER NAME FILTERING"> section for more details.
The values for the B<--warning> and B<--critical> options are amounts of
time, and default to '2 minutes' and '5 minutes' respectively. Valid units
are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or
abbreviated to just the first letter. If no units are given, the unit is
assumed to be seconds.
Example 1: Give a warning if any query has been running longer than 3 minutes, and a critical if longer than 5 minutes.
check_postgres_query_time --port=5432 --warning='3 minutes' --critical='5 minutes'
Example 2: Using default values (2 and 5 minutes), check all databases except those starting with 'template'.
check_postgres_query_time --port=5432 --exclude=~^template
Example 3: Warn if user 'don' has a query running over 20 seconds
check_postgres_query_time --port=5432 --inclucdeuser=don --warning=20s
=item B<txn_time> (symlink: C<check_postgres_txn_time>)
Checks the length of open transactions on one or more databases.
There is no need to run this command more than once per database cluster.
Databases can be filtered by use of the
B<--include> and B<--exclude> options. See the L</"BASIC FILTERING"> section
for more details. The owner of the transaction can also be filtered, by use of
the B<--includeuser> and B<--excludeuser> options.
See the L</"USER NAME FILTERING"> section for more details.
The values or the B<--warning> and B<--critical> options are units of time, and
must be provided (no default). Valid units are 'seconds', 'minutes', 'hours',
or 'days'. Each may be written singular or abbreviated to just the first letter.
If no units are given, the units are assumed to be seconds.
This action requires Postgres 8.3 or better.
Example 1: Give a critical if any transaction has been open for more than 10 minutes:
check_postgres_txn_time --port=5432 --critical='10 minutes'
Example 1: Warn if user 'warehouse' has a transaction open over 30 seconds
check_postgres_txn_time --port-5432 --warning=30s --includeuser=warehouse
=item B<txn_idle> (symlink: C<check_postgres_txn_idle>)
Checks the length of "idle in transaction" queries on one or more databases. There is
no need to run this more than once on the same database cluster. Databases can be filtered
by using the B<--include> and B<--exclude> options. See the L<"BASIC FILTERING">
section below for more details.
The B<--warning> and B<--critical> options are given as units of time, and both must
be provided (there are no defaults). Valid units are 'seconds', 'minutes', 'hours',
or 'days'. Each may be written singular or abbreviated to just the first letter.
If no units are given, the unit are assumed to be seconds.
This action requires Postgres 8.3 or better.
Example 1: Give a warning if any connection has been idle in transaction for more than 15 seconds:
check_postgres_txn_idle --port=5432 --warning='15 seconds'
=item B<rebuild_symlinks>
=item B<rebuild_symlinks_force>
This action requires no other arguments, and does not connect to any databases,
but simply creates symlinks in the current directory for each action, in the form
B<check_postgres_E<lt>action_nameE<gt>>.
If the file already exists, it will not be overwritten. If the action is rebuild_symlinks_force,
then symlinks will be overwritten.
=item B<settings_checksum> (symlink: C<check_postgres_settings_checksum>)
Check that all the Postgres settings are the same as last time you checked.
This is done by generating a checksum of a sorted list of setting names and
their values. Note that different users in the same database may have different
checksums, due to ALTER USER usage, and due to the fact that superusers see more
settings than ordinary users. Either the B<--warning> or the B<--critical> option
should be given, but not both. The value of each one is the checksum, a
32-character hexadecimal value. You can run with the special C<--critical=0> option
to find out an existing checksum.
This action requires the Digest::MD5 module.
Example 1: Find the initial checksum for the database on port 5555 using the default user (usually postgres)
check_postgres_settings_checksum --port=5555 --critical=0
Example 2: Make sure no settings have changed and warn if so, using the checksum from above.
check_postgres_settings_checksum --port=5555 --warning=cd2f3b5e129dc2b4f5c0f6d8d2e64231
=item B<timesync> (symlink: C<check_postgres_timesync>)
Compares the local system time with the time reported by one or more databases.
The B<--warning> and B<--critical> options represent the number of seconds between
the two systems before an alert is given. If neither is specified, the default values
are used, which are '2' and '5'. The warning value cannot be greater than the critical
value. Due to the non-exact nature of this test, values of '0' or '1' are not recommended.
The string returned shows the time difference as well as the time on each side written out.
Example 1: Check that databases on hosts ankh, morpork, and klatch are no more than 3 seconds off from the local time:
check_postgres_timesync --host=ankh,morpork.klatch --critical=3
=item B<txn_wraparound> (symlink: C<check_postgres_txn_wraparound>)
Checks how close to transaction wraparound one or more databases are getting.
The B<--warning> and B<--critical> options indicate the number of transactions
left, and must be a positive integer. If either option is not given, the default
values of 1.3 and 1.4 billion are used. There is no need to run this command
more than once per database cluster. For a more detailed discussion of what this
number represents and what to do about it, please visit the page
L<http://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND>
The warning and value can have underscores in the number for legibility, as Perl does.
Example 1: Check the default values for the localhost database
check_postgres_txn_wraparound --host=localhost
Example 2: Check port 6000 and give a critical at 1.7 billion transactions left:
check_postgres_txn_wraparound --port=6000 --critical=1_700_000_000t
=item B<wal_files> (symlink: C<check_postgres_wal_files>)
Checks how many WAL files exist in the F<pg_xlog> directory, which is found
off of your B<data_directory>, sometimes as a symlink to another physical disk for
performance reasons. This action must be run as a superuser, in order to access the
contents of the F<pg_xlog> directory. The minimum version to use this action is
Postgres 8.1. The B<--warning> and B<--critical> options are simply the number of
files in the F<pg_xlog> directory. What number to set this to will vary, but a general
guideline is to put a number slightly higher than what is normally there, to catch
problems early.
Normally, WAL files are closed and then re-used, but a long-running open
transaction, or a faulty B<archive_command> script, may cause Postgres to
create too many files. Ultimately, this will cause the disk they are on to run
out of space, at which point Postgres will shut down.
Example 1: Check that the number of WAL files is 20 or less on host "pluto"
check_postgres_txn_wraparound --host=pluto --critical=20
=item B<version> (symlink: C<check_version>)
Checks that the required version of Postgres is running. The
B<--warning> and B<--critical> options (only one is required) must be of
the format B<X.Y> or B<X.Y.Z> where B<X> is the major version number,
B<Y> is the minor version number, and B<Z> is the revision.
Example 1: Give a warning if the database on port 5678 is not version 8.4.10:
check_postgres_version --port=5678 -w=8.4.10
Example 2: Give a warning if any databases on hosts valley,grain, or sunshine is not 8.3:
check_postgres_version -H valley,grain,sunshine --critical=8.3
=back
=head1 BASIC FILTERING
The options B<--include> and B<--exclude> can be combined to limit which
things are checked, depending on the action. The name of the database can
be filtered when using the following actions:
backends, database_size, last_vacuum, last_analyze, locks, and query_time.
The name of a relation can be filtered when using the following actions:
bloat, index_size, table_size, and relation_size.
The name of a setting can be filtered when using the settings_checksum action.
The name of a file system can be filtered when using the disk_space action.
The name of a setting can be filtered when using the settings_checksum action.
If only an include option is given, then ONLY those entries that match will be
checked. However, if given both exclude and include, the exclusion is done first,
and the inclusion after, to reinstate things that may have been excluded. Both
B<--include> and B<--exclude> can be given multiple times,
and/or as comma-separated lists. A leading tilde will match the following word
as a regular expression.
Examples:
Only checks items named pg_class:
--include=pg_class
Only checks items containing the letters 'pg_':
--include=~pg_
Only check items beginning with 'pg_':
--include=~^pg_
Exclude the item named 'test':
--exclude=test
Exclude all items containing the letters 'test:
--exclude=~test
Exclude all items containing the letters 'ace', but allow the item 'faceoff':
--exclude=~ace --include=faceoff
Exclude all items which start with the letters 'pg_', which contain the letters 'slon',
or which are named 'sql_settings' or 'green'. Specifically check items with the letters 'prod' in their names, and always check the item named 'pg_relname':
--exclude=~^pg_,~slon,sql_settings --exclude=green --include=~prod,pg_relname
=head1 USER NAME FILTERING
The options B<--includeuser> and B<--excludeuser> can be used on some actions
to only examine database objects owned by (or not owned by) one or more users.
An B<--includeuser> option always trumps an B<--excludeuser> option. You can
give each option more than once for multiple users, or you can give a
comma-separated list. The actions that currently use these options are:
=over 4
=item database_size
=item last_analyze
=item last_vacuum
=item query_time
=item relation_size
=item txn_time
=back
Examples:
Only check items owned by the user named greg:
--includeuser=greg
Only check items owned by either watson or crick:
--includeuser=watson,crick
Only check items owned by crick,franklin, watson, or wilkins:
--includeuser=watson --includeuser=franklin --includeuser=crick,wilkins
Check all items except for those belonging to the user scott:
--excludeuser=scott
=head1 TEST MODE
To help in setting things up, this program can be run in a "test mode" by
specifying the B<--test> option. This will perform some basic tests to
make sure that the databases can be contacted, and that certain per-action
prerequisites are met. Currently, we check that the user is a superuser
if required by that action, and that the version of Postgres is new enough
for those actions that depend on a specific version.
=head1 DEPENDENCIES
Access to a working version of psql, and the following very standard Perl modules:
=over 4
=item B<Cwd>
=item B<Getopt::Long>
=item B<File::Basename>
=item B<File::Temp>
=item B<Time::HiRes> (if C<$opt{showtime}> is set to true, which is the default)
=back
The B<settings_checksum> action requires the B<Digest::MD5> module.
Some actions require access to external programs. If psql is not explicitly
specified, the command B<which> is used to find it. The program B</bin/df>
is needed by the B<check_disk_space> action.
=head1 DEVELOPMENT
Development happens using the git system. You can clone the latest version by doing:
git-clone http://bucardo.org/check_postgres.git
=head1 HISTORY
Items not specifically attributed are by Greg Sabino Mullane.
=over 4
=item B<Version 1.5.2> (May 2, 2008)
Fix problem with too eager creation of custom pgpass file.
=item B<Version 1.5.1> (April 17, 2008)
Add example Nagios configuration settings (Brian A. Seklecki)
=item B<Version 1.5.0> (April 16, 2008)
Add the --includeuser and --excludeuser options. Documentation cleanup.
=item B<Version 1.4.3> (April 16, 2008)
Add in the 'output' concept for future support of non-Nagios programs.
=item B<Version 1.4.2> (April 8, 2008)
Fix bug preventing --dbpass argument from working (Robert Treat).
=item B<Version 1.4.1> (April 4, 2008)
Minor documentation fixes.
=item B<Version 1.4.0> (April 2, 2008)
Have check_wal_files use pg_ls_dir (idea by Robert Treat).
For last_vacuum and last_analyze, respect autovacuum effects, add separate
autovacuum checks (ideas by Robert Treat).
=item B<Version 1.3.1> (April 2, 2008)
Have txn_idle use query_start, not xact_start.
=item B<Version 1.3.0> (March 23, 2008)
Add in txn_idle and txn_time actions.
=item B<Version 1.2.0> (February 21, 2008)
Add the check_wal_files method, which counts the number of WAL files
in your pg_xlog directory.
Fix some typos in the docs.
Explicitly allow -v as an argument.
Allow for a null syslog_facility in check_logfile.
=item B<Version 1.1.2> (February 5, 2008)
Fix error preventing --action=rebuild_symlinks from working.
=item B<Version 1.1.1> (February 3, 2008)
Switch vacuum and analyze date output to use 'DD', not 'D'. (Glyn Astill)
=item B<Version 1.1.0> (December 16, 2008)
Fixes, enhancements, and performance tracking.
Add performance data tracking via --showperf and --perflimit
Lots of refactoring and cleanup of how actions handle arguments.
Do basic checks to figure out syslog file for 'logfile' action.
Allow for exact matching of beta versions with 'version' action.
Redo the default arguments to only populate when neither 'warning' nor 'critical' is provided.
Allow just warning OR critical to be given for the 'timesync' action.
Remove 'redirect_stderr' requirement from 'logfile' due to 8.3 changes.
Actions 'last_vacuum' and 'last_analyze' are 8.2 only (Robert Treat)
=item B<Version 1.0.16> (December 7, 2007)
First public release, December 2007
=back
=head1 BUGS AND LIMITATIONS
The index bloat size optimization is still very rough.
Some actions may not work on older versions of Postgres (before 8.0).
Please report any problems to greg@endpoint.com.
=head1 AUTHOR
Greg Sabino Mullane <greg@endpoint.com>
=head1 NAGIOS EXAMPLES
Some example Nagios configuration settings using this script:
define command {
command_name check_postgres_size
command_line $USER2$/check_postgres.pl -H $HOSTADDRESS$ -u pgsql -db postgres --action database_size -w $ARG1$ -c $ARG2$
}
define command {
command_name check_postgres_locks
command_line $USER2$/check_postgres.pl -H $HOSTADDRESS$ -u pgsql -db postgres --action locks -w $ARG1$ -c $ARG2$
}
define service {
use generic-other
host_name dbhost.gtld
service_description dbhost PostgreSQL Service Database Usage Size
check_command check_postgres_size!256000000!512000000
}
define service {
use generic-other
host_name dbhost.gtld
service_description dbhost PostgreSQL Service Database Locks
check_command check_postgres_locks!2!3
}
=head1 LICENSE AND COPYRIGHT
Copyright (c) 2007-2008 Greg Sabino Mullane <greg@endpoint.com>.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED
WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
OF SUCH DAMAGE.
=cut