kurtrottmann / check-linux-by-ssh

Shinken checks designed to be run without installing something on the distant linux (aside ssh keys of course!)

io-shinken-checks-linux

Specifics checks for linux based on pure ssh polling, with nothing to install on the target

Dependencies

• python
• python-paramiko

Connexion

This check is for checking the SSH connexion to the distant server.

$ check_ssh_connexion.py -H localhost -u shinken

Memory

The Memory check is done by the check_memory_by_ssh.py script.

$ check_memory_by_ssh.py -H localhost -u shinken -w "75%" -c "90%"

Uptime

The Uptime check is done by the check_uptime_by_ssh.py script. It only take a -c option, the number of second : below it's critical, higher it's ok. By default it's 3600s.

$ check_uptime_by_ssh.py -H localhost -u shinken -c 3600

NTP Sync

The NTP sync check is done by the check_ntp_sync_by_ssh.py script. It will go in warning if no ntp server is the reference, and -w/-c options will set the maximum delay values.

$ check_ntp_sync_by_ssh.py -H localhost -u shinken -w 10 -c 60

Processes (Memory)

Look at the memory of a process, or a pack of processes. It's done by the check_processes_by_ssh script

Look that the sum of all chrome processes are not over 700 or 800MB $check_processes_by_ssh.py -H localhost -u shinken -C chrome -w 700 -c 800 -S

Look for each chrome processe if they are not over 100 or 200MB $check_processes_by_ssh.py -H localhost -u shinken -C chrome -w 100 -c 200

Look for all process, and warn if one is over 100/200MB $check_processes_by_ssh.py -H localhost -u shinken -w 100 -c 200

Disks

The Disks check is done by the check_disks_by_ssh.py script.

$ check_disks_by_ssh.py -H localhost -u shinken -w "75%" -c "90%"

Load average

The load average values are checks with the check_load_average_by_ssh script. There are two modes : strict values, and cpu based values. 'default : strict)

Will warn if the load average is highen than 1 or 2 $ check_load_average_by_ssh.py -H localhost -u shinken -w 1,1,1 -c 2,2,2

Will warn if the load average is highen than 1nb_cpus or 2nb_cpus $ check_load_average_by_ssh.py -H localhost -u shinken -w 1,1,1 -c 2,2,2 -C

CPU activities

The cpu states are checks by the check_cpu_stats.py script. There is no warning or critical values need here.

$ check_cpu_stats.py -H localhost -u shinken

DISKS activities

The disks I/O are checked by the check_disks_stats_by_ssh.py. No warning nor critical values need.

$ check_disks_stats_by_ssh.py -H localhost -u shinken

TCP states

The TCP states are checked by the check_tcp_states_by_ssh.py plugin. No warning nor critical values need.

$ check_tcp_states_by_ssh.py - H localhost -u shinken

KERNEL stats

The KERNEL states are checked by the check_kernel_stats_by_ssh.py plugin. No warning nor critical values need.

$ check_kernel_stats_by_ssh.py - H localhost -u shinken

NFS stats

The NFS states are checked by the check_nfs_stats_by_ssh.py plugin. No warning nor critical values need.

$ check_nfs_stats_by_ssh.py - H localhost -u shinken

Interface activities

The network activity is checked by the check_net_stats_by_ssh.py plugin. No need for warning nor critical.

$ check_net_stats_by_ssh.py - H localhost -u shinken

Read Only file systems

The file system mount are checks. If a FS is in read only, it will raise a critical error.

$ check_ro_filesystem_by_ssh.py - H localhost -u shinken