This package contains Nagios plugins for monitoring DRBD resources. The plugins use output from the following sources to determine the state of each resource:
/proc/drbd/usr/sbin/drbdadm sh-resources/usr/sbin/drbdetup events2 --now --statistics
The following DRBD kernel modules and DRBD Utilities are supported:
- DRBD 8.4.x with drbd-utils 8.9.6
- DRBD 9.0.x with drbd-utils 8.9.6
This package contains the following scripts:
- check_drbd9.pl - Checks DRBD resources
- dump_drbd9.pl - Dumps resource information (used for debugging)
check_drbd9.pl:
Usage: check_drbd9.pl [OPTIONS]
OPTIONS:
-c state change specified state to 'CRIT' (example: SyncSource)
-d pattern same as '-i', added for compatibility with legacy check
-h display this message
-i pattern include resource name or resource minor (default: all)
-l list all OKAY resources after CRIT and WARN resources
-o state change specified state to 'OKAY' (example: StandAlone)
-q quiet output
-t display terse details
-V display program version
-v display OKAY resources
-w state change specified state to 'WARN' (example: SyncTarget)
-x pattern exclude resource name or resource minor
ROLE STATES:
primary (OKAY) secondary (OKAY) unconfigured (CRIT)
unknown (WARN)
LOCAL DISK STATES:
attaching (WARN) consistent (OKAY) detaching (WARN)
diskless (CRIT) dunknown (WARN) failed (CRIT)
inconsistent (CRIT) negotiating (WARN) outdated (CRIT)
uptodate (OKAY)
CONNECTION STATES:
brokenpipe (CRIT) connected (OKAY) connecting (WARN)
disconnecting (WARN) networkfailure (CRIT) protocolerror (CRIT)
standalone (WARN) teardown (WARN) timeout (CRIT)
unconnected (CRIT) wfconnection (CRIT) wfreportparams (CRIT)
PEER DISK STATES:
attaching (WARN) consistent (OKAY) detaching (WARN)
diskless (WARN) dunknown (WARN) failed (WARN)
inconsistent (WARN) negotiating (WARN) outdated (WARN)
uptodate (OKAY)
REPLICATION STATES:
ahead (WARN) behind (CRIT) established (OKAY)
off (WARN) pausedsyncs (CRIT) pausedsynct (CRIT)
startingsyncs (WARN) startingsynct (WARN) syncsource (WARN)
synctarget (WARN) verifys (WARN) verifyt (WARN)
wfbitmaps (CRIT) wfbitmapt (CRIT) wfsyncuuid (WARN)
syzdek@hypervisor$ ./check_drbd9.pl -o StandAlone -d all -l
DRBD: 1 warn, 20 okay|
-
version: 8.4.7-1 (api:1/proto:86-101)
resources: 21
-
lab.prv.ldaptest23 role:Secondary (WARN)
vol:0 disk:UpToDate minor:18 size:32766MB
vol:1 disk:UpToDate minor:19 size:65533MB
peer role:Unknown conn:Connecting
.. vol:0 peer:DUnknown repl:Off sync:100.00%
.. vol:1 peer:DUnknown repl:Off sync:100.00%
-
lab.prv.cbrown role:Secondary (OKAY)
lab.prv.dnstest20 role:Secondary (OKAY)
lab.prv.dnstest21 role:Secondary (OKAY)
lab.prv.dnstest22 role:Secondary (OKAY)
lab.prv.dnstest23 role:Secondary (OKAY)
lab.prv.dnstestlvs20 role:Secondary (OKAY)
lab.prv.dnstestlvs21 role:Secondary (OKAY)
lab.prv.drbdtest20 role:Secondary (OKAY)
lab.prv.drbdtest21 role:Secondary (OKAY)
lab.prv.kohfield role:Secondary (OKAY)
lab.prv.kshymkiw role:Secondary (OKAY)
lab.prv.ldaptest20 role:Secondary (OKAY)
lab.prv.ldaptest21 role:Secondary (OKAY)
lab.prv.ldaptest22 role:Secondary (OKAY)
lab.prv.ldaptestlvs20 role:Secondary (OKAY)
lab.prv.ldaptestlvs21 role:Secondary (OKAY)
lab.prv.ldaptestmirror01 role:Secondary (OKAY)
lab.prv.ldaptestmirror02 role:Secondary (OKAY)
lab.prv.simestd role:Secondary (OKAY)
lab.prv.wallette role:Secondary (OKAY)
|
syzdek@hypervisor$
Another example:
syzdek@hypervisor$ ./check_drbd9.pl -o StandAlone -d all -l
DRBD: 1 warn, 20 okay|
-
version: 9.0.2-1 (api:2/proto:86-111)
transports: (api:14): tcp (1.0.0)
resources: 21
-
lab.prv.drbdtest20 role:Primary (WARN)
vol:0 disk:UpToDate minor:30 size:32766MB
xen67 role:Secondary conn:Connected
.. vol:0 peer:Inconsistent repl:SyncSource sync:11.15%
-
lab.prv.cbrown role:Primary (OKAY)
lab.prv.dnstest20 role:Primary (OKAY)
lab.prv.dnstest21 role:Secondary (OKAY)
lab.prv.dnstest22 role:Primary (OKAY)
lab.prv.dnstest23 role:Secondary (OKAY)
lab.prv.dnstestlvs20 role:Primary (OKAY)
lab.prv.dnstestlvs21 role:Secondary (OKAY)
lab.prv.drbdtest21 role:Secondary (OKAY)
lab.prv.kohfield role:Secondary (OKAY)
lab.prv.kshymkiw role:Primary (OKAY)
lab.prv.ldaptest20 role:Primary (OKAY)
lab.prv.ldaptest21 role:Primary (OKAY)
lab.prv.ldaptest22 role:Secondary (OKAY)
lab.prv.ldaptest23 role:Secondary (OKAY)
lab.prv.ldaptestlvs20 role:Primary (OKAY)
lab.prv.ldaptestlvs21 role:Secondary (OKAY)
lab.prv.ldaptestmirror01 role:Primary (OKAY)
lab.prv.ldaptestmirror02 role:Secondary (OKAY)
lab.prv.simestd role:Primary (OKAY)
lab.prv.wallette role:Secondary (OKAY)
|
syzdek@hypervisor$
Example configuration object for Icinga 2:
object CheckCommand "check_drbd9" {
import "plugin-check-command"
command = [ PluginDir + "/check_drbd9" ]
arguments = {
"-c" = {
value = "$drbd9_crit_states$"
description = "change specified state to 'CRIT' (example: SyncSource)"
}
"-w" = {
value = "$drbd9_warn_states$"
description = "change specified state to 'WARN' (example: SyncTarget)"
}
"-o" = {
value = "$drbd9_okay_states$"
description = "change specified state to 'OKAY' (example: StandAlone)"
}
"-i" = {
value = "$drbd9_include$"
description = "include resource name or resource minor (default: all)"
}
"-x" = {
value = "$drbd9_exclude$"
description = "exclude resource name or resource minor"
}
"-l" = {
set_if = "$drbd9_list_all$"
description = "list all OKAY resources after CRIT and WARN resources"
}
"-v" = {
set_if = "$drbd9_okay$"
description = "display OKAY resources"
}
"-q" = {
set_if = "$drbd9_quiet$"
description = "quiet output"
}
"-t" = {
set_if = "$drbd9_terse$"
description = "display terse details"
}
}
}
Example command object configurations:
# Check all resources, treat StandAlone as OKAY
define command{
command_name check_drbd
command_line $USER1$/check_drbd9.pl -o StandAlone -i all -l
}
By default, all resources are checked.
DRBD service checks:
define service{
use generic-service
host_name hypervisorA.foo.org
display_name DRBD
service_description DRBD
check_command check_drbd
}
NRPE Configuration:
nrpe_user=nagios
nrpe_group=nagios
allowed_hosts=10.0.0.0/8,127.0.0.0/8,172.16.0.0/12,192.168.0.0/16
dont_blame_nrpe=0
allow_bash_command_substitution=0
command[check_drbd]=/usr/libexec/nagios/check_drbd -d All -o StandAlone
Due to buffer limitations of NRPE (1024 bytes as of this writting), it is
recommended that the -l option not be used on systems with more than than
3-4 resources. If CRIT and WARN messages are being truncated, enable terse
output with -t to print each one line messages for CRIT and WARN states.
Nagios Configuration:
define command{
command_name check_nrpe
command_line $USER1$/check_nrpe -u -H $HOSTADDRESS$ -c $ARG1$
}
define service{
use generic-service
host_name hypervisorA.foo.org
display_name DRBD
service_description DRBD
check_command check_nrpe!check_drbd
}