nagios-plugins

just some basic nagios plugin

ZFS check for linux

example output

# /usr/lib/nagios/plugins/check_zfs # one pool
OK storage-02(ONLINE 10.3T/87T),storage-03(ONLINE 75.6T/87T),storage-81(ONLINE 1.00T/87T)
# resilver example
WARNING storage-02(ONLINE 11.1T/87T),storage-03(RESILVER 18.53%,223M/s,82h51m),storage-81(ONLINE 1.00T/87T)
# /usr/lib/nagios/plugins/check_zfs # multiple storage pool
OK storage-03(ONLINE 118T/390T)

example cronjob

* * * * * root /usr/bin/perl /usr/lib/nagios/plugins/check_zfs |xargs echo $(hostname) pass_Raid| xargs passive_check_send

integrating with NRPE

The Nagios Remote Plugin Executor can run this script upon request from a Nagios monitor server. The NRPE daemon runs as an unprivileged user (typically named nrpe or nagios) although the zfs/zpool commands require superuser privileges.

Included is an example configuration file that can be placed in /etc/sudoers.d'. It will allow any non-root user to run the "read-only" zfsandzpool` commands while preventing the execution of commands that modify filesystems and storage pools.

A typical NRPE configuration files is placed in /etc/nrpe.d on RHEL-derived platforms and /etc/nagios/nrpe.d on Debian derivatives. These files point to nagios plugins for execution. For example

command[check_zpool]=/usr/lib/nagios/plugins/check_zpool

On RHEL derivatives, the plugins are stored under /usr/lib64. The nagios monitor must then be configured to check the host using this command. The example below is not exact but intended to provide the outline. Please examine nagios documentation for the full range of nagios object definitions and their options.

define service {
  use                    generic-service
  service_description    zpool-health
  hostgroup_name         storage
  ...
  check_command          check_nrpe_1arg!check_zpool
}

Check open filehandles (comparing open filehandles with proc's rlimit nofile) - linux only

/usr/lib/nagios/plugins/check_procs_fh [-w warningpercent] [-c criticalpercent]

example output

# /usr/lib/nagios/plugins/check_procs_fh -w 1 -c 2
pass_ProcFD CRITICAL: (W:sshd;3148;11/1024), (C:syslog-ng;7486;2004/65536), (W:udevd;1529;11/1024), (W:lldpd;7063;14/1024), (W:rpcbind;6170;12/1024), (C:rpc.mountd;16466;21/1024)

example cronjob

same as zfs

Check bacula job status - SQL

/usr/lib/nagios/plugins/check_bacula

example output

# /usr/lib/nagios/plugins/check_bacula
pass_Bacula CRITICAL: Success: 75, Failed: 3 (Job1AutoMysqlBackupDaily, Job2AutoMysqlBackupDaily, Job3BackupDaily)

example cronjob

same as zfs