enh(fortigate/snmp): mode cluster-status - add cpu and memory metrics (#3386)

This commit is contained in:
qgarnier 2022-01-07 11:02:13 +01:00 committed by GitHub
parent 8e246a6c3e
commit 8f1021295d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 119 additions and 75 deletions

View File

@ -24,12 +24,12 @@ use base qw(centreon::plugins::templates::counter);
use strict;
use warnings;
use centreon::plugins::templates::catalog_functions qw(catalog_status_threshold);
use centreon::plugins::templates::catalog_functions qw(catalog_status_threshold_ng);
use Digest::MD5 qw(md5_hex);
sub custom_status_output {
my ($self, %options) = @_;
return sprintf(
"status is '%s' [hostname: %s] [role: %s] [checksum: %s]",
$self->{result_values}->{sync_status},
@ -39,6 +39,30 @@ sub custom_status_output {
);
}
sub prefix_global_output {
my ($self, %options) = @_;
return 'Nodes ';
}
sub node_long_output {
my ($self, %options) = @_;
return sprintf(
"checking node '%s'",
$options{instance_value}->{serial}
);
}
sub prefix_node_output {
my ($self, %options) = @_;
return sprintf(
"node '%s' ",
$options{instance_value}->{serial}
);
}
sub custom_status_calc {
my ($self, %options) = @_;
@ -56,32 +80,27 @@ sub custom_status_calc {
return 0;
}
sub prefix_status_output {
my ($self, %options) = @_;
return "Node '" . $options{instance_value}->{serial} . "' ";
}
sub prefix_global_output {
my ($self, %options) = @_;
return 'Nodes ';
}
sub set_counters {
my ($self, %options) = @_;
$self->{maps_counters_type} = [
{ name => 'global', type => 0, cb_prefix_output => 'prefix_global_output' },
{ name => 'nodes', type => 1, cb_prefix_output => 'prefix_status_output', message_multiple => 'All cluster nodes status are ok' }
{ name => 'nodes', type => 3, cb_prefix_output => 'prefix_node_output', cb_long_output => 'node_long_output',
indent_long_output => ' ', message_multiple => 'All cluster nodes are ok',
group => [
{ name => 'status', type => 0, skipped_code => { -10 => 1 } },
{ name => 'cpu', type => 0, skipped_code => { -10 => 1 } },
{ name => 'memory', type => 0, skipped_code => { -10 => 1 } }
]
}
];
$self->{maps_counters}->{global} = [
{ label => 'total-nodes', nlabel => 'cluster.nodes.total.count', display_ok => 0, set => {
{ label => 'total-nodes', nlabel => 'cluster.nodes.count', set => {
key_values => [ { name => 'total_nodes' } ],
output_template => 'total nodes: %d',
output_template => 'total: %d',
perfdatas => [
{ label => 'total_nodes', template => '%d', min => 0 }
{ template => '%d', min => 0 }
]
}
},
@ -89,35 +108,58 @@ sub set_counters {
key_values => [ { name => 'synchronized' } ],
output_template => 'synchronized: %d',
perfdatas => [
{ label => 'synchronized_nodes', template => '%d', min => 0 }
{ template => '%d', min => 0 }
]
}
},
{ label => 'not-synchronized', nlabel => 'cluster.nodes.notsynchronized.count',set => {
key_values => [ { name => 'not_synchronized' } ],
output_template => 'not synchronized: %d',
{ label => 'unsynchronized', nlabel => 'cluster.nodes.unsynchronized.count',set => {
key_values => [ { name => 'unsynchronized' } ],
output_template => 'unsynchronized: %d',
perfdatas => [
{ label => 'not_synchronized_nodes', template => '%d', min => 0 }
{ template => '%d', min => 0 }
]
}
},
{ label => 'total-checksums', nlabel => 'cluster.checksums.total.count', display_ok => 0, set => {
key_values => [ { name => 'total_checksums' } ],
output_template => 'total checksums: %d',
output_template => 'checksums: %d',
perfdatas => [
{ label => 'total_checksums', template => '%d', min => 0 }
{ template => '%d', min => 0 }
]
}
}
];
$self->{maps_counters}->{nodes} = [
{ label => 'status', threshold => 0, set => {
$self->{maps_counters}->{status} = [
{ label => 'status', type => 2, critical_default => '%{role} ne %{roleLast} or %{sync_status} =~ /unsynchronized/', set => {
key_values => [ { name => 'serial' }, { name => 'hostname' }, { name => 'sync_status' }, { name => 'role' }, { name => 'checksum' } ],
closure_custom_calc => \&custom_status_calc,
closure_custom_output => $self->can('custom_status_output'),
closure_custom_perfdata => sub { return 0; },
closure_custom_threshold_check => \&catalog_status_threshold
closure_custom_threshold_check => \&catalog_status_threshold_ng
}
}
];
$self->{maps_counters}->{cpu} = [
{ label => 'cpu-utilization', nlabel => 'node.cpu.utilization.percentage', set => {
key_values => [ { name => 'cpu_usage' }, { name => 'serial' } ],
output_template => 'cpu usage: %.2f %%',
perfdatas => [
{ template => '%.2f', unit => '%', min => 0, max => 100, label_extra_instance => 1, instance_use => 'serial' }
]
}
}
];
$self->{maps_counters}->{memory} = [
{ label => 'memory-usage', nlabel => 'node.memory.usage.percentage', set => {
key_values => [ { name => 'memory_usage' }, { name => 'serial' } ],
output_template => 'memory used: %.2f %%',
output_change_bytes => 1,
perfdatas => [
{ template => '%.2f', unit => '%', min => 0, max => 100, label_extra_instance => 1, instance_use => 'serial' }
]
}
}
];
@ -125,43 +167,34 @@ sub set_counters {
sub new {
my ($class, %options) = @_;
my $self = $class->SUPER::new(package => __PACKAGE__, %options, statefile => 1);
my $self = $class->SUPER::new(package => __PACKAGE__, %options, statefile => 1, force_new_perfdata => 1);
bless $self, $class;
$options{options}->add_options(arguments => {
'warning-status:s' => { name => 'warning_status', default => '' },
'critical-status:s' => { name => 'critical_status', default => '%{role} ne %{roleLast} or %{sync_status} =~ /not synchronized/' },
'one-node-status:s' => { name => 'one_node_status' } # not used, use --opt-exit instead
});
return $self;
}
sub check_options {
my ($self, %options) = @_;
$self->SUPER::check_options(%options);
$self->change_macros(macros => ['warning_status', 'critical_status']);
}
my $map_ha_mode = {
1 => 'standalone',
2 => 'activeActive',
3 => 'activePassive'
};
my $map_sync_status = {
0 => 'not synchronized',
0 => 'unsynchronized',
1 => 'synchronized'
};
my $mapping = {
fgHaStatsSerial => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.2' },
fgHaStatsHostname => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.11' },
fgHaStatsSyncStatus => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.12', map => $map_sync_status },
fgHaStatsGlobalChecksum => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.15' },
fgHaStatsMasterSerial => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.16' }
cpuUsage => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.3' }, # fgHaStatsCpuUsage
memUsage => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.4' }, # fgHaStatsMemUsage
hostname => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.11' }, # fgHaStatsHostname
syncStatus => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.12', map => $map_sync_status }, # fgHaStatsSyncStatus
globalChecksum => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.15' }, # fgHaStatsGlobalChecksum
masterSerial => { oid => '.1.3.6.1.4.1.12356.101.13.2.1.1.16' } # fgHaStatsMasterSerial
};
my $oid_fgHaStatsEntry = '.1.3.6.1.4.1.12356.101.13.2.1.1';
my $oid_fgHaSystemMode = '.1.3.6.1.4.1.12356.101.13.1.1.0';
sub manage_selection {
@ -174,41 +207,52 @@ sub manage_selection {
$self->{output}->option_exit();
}
$self->{output}->output_add(short_msg => 'HA mode: ' . $map_ha_mode->{ $mode->{$oid_fgHaSystemMode} });
$self->{output}->output_add(short_msg => 'High-availibility mode: ' . $map_ha_mode->{ $mode->{$oid_fgHaSystemMode} });
my $oid_serial = '.1.3.6.1.4.1.12356.101.13.2.1.1.2'; # fgHaStatsSerial
my $snmp_result = $options{snmp}->get_table(
oid => $oid_fgHaStatsEntry,
oid => $oid_serial,
nothing_quit => 1
);
$self->{global} = { synchronized => 0, not_synchronized => 0, total_nodes => 0 };
$self->{global} = { synchronized => 0, unsynchronized => 0, total_nodes => 0 };
$self->{nodes} = {};
my $checksums = {};
foreach my $oid (keys %$snmp_result) {
next if ($oid !~ /^$mapping->{fgHaStatsSerial}->{oid}\.(.*)$/);
foreach (keys %$snmp_result) {
/^$oid_serial\.(.*)$/;
my $instance = $1;
my $result = $options{snmp}->map_instance(mapping => $mapping, results => $snmp_result, instance => $instance);
$self->{nodes}->{$instance} = { serial => $snmp_result->{$_} };
}
$checksums->{$result->{fgHaStatsGlobalChecksum}} = 1;
$self->{nodes}->{$instance} = {
serial => $result->{fgHaStatsSerial},
hostname => $result->{fgHaStatsHostname},
sync_status => $result->{fgHaStatsSyncStatus},
role => ($result->{fgHaStatsMasterSerial} eq '' || $result->{fgHaStatsMasterSerial} =~ /$result->{fgHaStatsSerial}/) ? 'master' : 'slave',
checksum => $result->{fgHaStatsGlobalChecksum}
return if (scalar(keys %{$self->{nodes}}) <= 0);
$options{snmp}->load(
oids => [ map($_->{oid}, values(%$mapping)) ],
instances => [ map($_, keys(%{$self->{nodes}})) ],
instance_regexp => '^(.*)$'
);
$snmp_result = $options{snmp}->get_leef();
my $checksums = {};
foreach (keys %{$self->{nodes}}) {
my $result = $options{snmp}->map_instance(mapping => $mapping, results => $snmp_result, instance => $_);
$checksums->{ $result->{globalChecksum} } = 1;
$self->{nodes}->{$_}->{status} = {
serial => $self->{nodes}->{$_}->{serial},
hostname => $result->{hostname},
sync_status => $result->{syncStatus},
role => ($result->{masterSerial} eq '' || $result->{masterSerial} =~ /$self->{nodes}->{$_}->{serial}/) ? 'master' : 'slave',
checksum => $result->{globalChecksum}
};
$result->{fgHaStatsSyncStatus} =~ s/ /_/;
$self->{global}->{$result->{fgHaStatsSyncStatus}}++;
$self->{nodes}->{$_}->{cpu} = { cpu_usage => $result->{cpuUsage}, serial => $self->{nodes}->{$_}->{serial} };
$self->{nodes}->{$_}->{memory} = { memory_usage => $result->{memUsage}, serial => $self->{nodes}->{$_}->{serial} };
$self->{global}->{ $result->{syncStatus} }++;
$self->{global}->{total_nodes}++;
}
$self->{global}->{total_checksums} = scalar(keys %$checksums);
if (scalar(keys %{$self->{nodes}}) <= 0) {
$self->{output}->add_option_msg(short_msg => 'No cluster nodes found');
$self->{output}->option_exit();
}
$self->{cache_name} = 'fortinet_fortigate_' . $self->{mode} . '_' . $options{snmp}->get_hostname() . '_' . $options{snmp}->get_port() . '_' .
(defined($self->{option_results}->{filter_counters}) ? md5_hex($self->{option_results}->{filter_counters}) : md5_hex('all'));
@ -220,26 +264,26 @@ __END__
=head1 MODE
Check cluster status (FORTINET-FORTIGATE-MIB).
Check cluster status.
=over 8
=item B<--warning-*> B<--critical-*>
Set thresholds.
Can be: 'total-nodes', 'synchronized', 'not-synchronized',
'total-checksums'.
=item B<--warning-status>
Set warning threshold for status (Default: '').
Set warning threshold for status.
Can used special variables like: %{serial}, %{hostname}, %{sync_status}, %{role}, %{roleLast}
=item B<--critical-status>
Set critical threshold for status (Default: '%{role} ne %{roleLast} or %{sync_status} !~ /synchronized/').
Set critical threshold for status (Default: '%{role} ne %{roleLast} or %{sync_status} =~ /unsynchronized/').
Can used special variables like: %{serial}, %{hostname}, %{sync_status}, %{role}, %{roleLast}
=item B<--warning-*> B<--critical-*>
Set thresholds.
Can be: 'total-nodes', 'synchronized', 'unsynchronized',
'total-checksums', 'cpu-utilization', 'memory-usage'.
=back
=cut