check_synology.py

#!/usr/bin/env python3

import argparse
import sys
import math
import re

import easysnmp

AUTHOR = "Frederic Werner"
VERSION = "1.1.0"

parser = argparse.ArgumentParser()
parser.add_argument("hostname", help="the hostname", type=str)
parser.add_argument("username", help="the snmp user name", type=str)
parser.add_argument("authkey", help="the auth key", type=str)
parser.add_argument("privkey", help="the priv key", type=str)
parser.add_argument("mode", help="the mode", type=str, choices=["load", "memory", "disk", "storage", "update", "status"])
parser.add_argument("-w", help="warning value for selected mode", type=int)
parser.add_argument("-c", help="critical value for selected mode", type=int)
parser.add_argument("-p", help="the snmp port", type=int, dest="port", default=161)
parser.add_argument("-e", help="SNMP privacy protocol encryption", type=str, default="AES128", choices=["AES128", "DES"])
parser.add_argument("-t", help="timeout for snmp connection", type=int, default=10)
parser.add_argument("-r", help="retries for snmp connection if timeout occurs", type=int, default=3)
args = parser.parse_args()

hostname = args.hostname
port = args.port
user_name = args.username
auth_key = args.authkey
priv_key = args.privkey
mode = args.mode
warning = args.w
critical = args.c
priv_protocol = args.e
snmp_timeout = args.t
snmp_retries = args.r

state = 'OK'

def croak(message=None):
    """
    Exit program with `UNKNOWN` state and error message.
    """
    global state
    state = 'UNKNOWN'
    message = message and str(message) or "unknown error"
    print('%s - %s' % (state, message))
    exitCode()

try:
    session = easysnmp.Session(
        hostname=hostname,
        remote_port=port,
        version=3,
        timeout=snmp_timeout,
        retries=snmp_retries,
        security_level="auth_with_privacy",
        security_username=user_name,
        auth_password=auth_key,
        auth_protocol="MD5",
        privacy_password=priv_key,
        privacy_protocol=priv_protocol)

except Exception as e:
    croak(e)

def snmpget(oid):
    """
    Return value from single OID.
    """
    try:
        res = session.get(oid)
        return res.value
    except Exception as e:
        croak(e)

def snmpwalk(oid):
    """
    Walk the given OID and return all child OIDs as a list of tuples of OID and value.
    """
    res = []
    try:
        res = session.walk(oid)
    except Exception as e:
        croak(e)
    return res

def exitCode():
    if state == 'OK':
        sys.exit(0)
    if state == 'WARNING':
        sys.exit(1)
    if state == 'CRITICAL':
        sys.exit(2)
    if state == 'UNKNOWN':
        sys.exit(3)

if mode == 'load':
    load1 = str(float(snmpget('1.3.6.1.4.1.2021.10.1.5.1'))/100)
    load5 = str(float(snmpget('1.3.6.1.4.1.2021.10.1.5.2'))/100)
    load15 = str(float(snmpget('1.3.6.1.4.1.2021.10.1.5.3'))/100)

    if warning and warning < int(math.ceil(float(load1))):
        state = 'WARNING'
    if critical and critical < int(math.ceil(float(load1))):
        state = 'CRITICAL'

    print(state + ' - load average: %s, %s, %s' % (load1, load5, load15), '| load1=%sc' % load1, 'load5=%sc' % load5, 'load15=%sc' % load15)
    exitCode()

if mode == 'memory':
    memory_total = float(snmpget('1.3.6.1.4.1.2021.4.5.0'))
    memory_unused = float(snmpget('1.3.6.1.4.1.2021.4.6.0'))
    memory_cached = float(snmpget('1.3.6.1.4.1.2021.4.15.0'))
    memory_usable = memory_unused + memory_cached
    memory_percent = 100 / memory_total * memory_usable

    if warning and warning > int(memory_percent):
        state = 'WARNING'
    if critical and critical > int(memory_percent):
        state = 'CRITICAL'

    print(state + ' - {:0.1f}% '.format(memory_percent) + 'usable ({0:0.1f} MB free and {1:0.1f} MB cached out of {2:0.1f} MB)'.format((memory_unused / 1024), (memory_cached / 1024), (memory_total / 1024)), '|memory_total=%dc' % memory_total, 'memory_unused=%dc' % memory_unused, 'memory_cached=%dc' % memory_cached, 'memory_usable=%dc' % memory_usable, 'memory_percent=%d' % memory_percent + '%')
    exitCode()

if mode == 'disk':
    """
    Synology Disk MIB (OID: .1.3.6.1.4.1.6574.2)

    OK:       Status from all disks is "Normal" and no temperature threshold is raised.
    WARNING:  Temperature threshold for warning level is reached on any disk.
    CRITICAL: Either the status from any disk is "SystemPartitionFailed" or "Crashed",
              or the temperature threshold for criticality level is reached on any disk.
    UNKNOWN:  No disk states collected via SNMP at all.
    """

    # 1. Retrieve and decode system metrics.
    maxDisk = 0
    states = []
    output = ''
    perfdata = '|'
    for item in snmpwalk('1.3.6.1.4.1.6574.2.1.1.2'):
        i = item.oid_index or item.oid.split('.')[-1]
        disk_name = item.value
        disk_status_nr = snmpget('1.3.6.1.4.1.6574.2.1.1.5.' + str(i))
        disk_health_status_nr = snmpget('1.3.6.1.4.1.6574.2.1.1.13.' + str(i))
        disk_temp = snmpget('1.3.6.1.4.1.6574.2.1.1.6.' + str(i))
        status_translation = {
            '1': "Normal",
            '2': "Initialized",
            '3': "NotInitialized",
            '4': "SystemPartitionFailed",
            '5': "Crashed"
        }
        health_status_translation = {
            '1': "Normal",
            '2': "Warning",
            '3': "Critical",
            '4': "Failing"
        }
        disk_status = status_translation.get(disk_status_nr)
        disk_health_status = health_status_translation.get(disk_health_status_nr)
        disk_name = disk_name.replace(" ", "")

        # 2. Compute textual and perfdata output.
        output += ' - ' + disk_name + ': Status: ' + disk_status + ', Temperature: ' + disk_temp + ' C' + ', Health status: ' + disk_health_status
        perfdata += 'temperature' + disk_name + '=' + disk_temp + 'c '

        # 3. Collect outcome for individual sensor state.

        # When state is already "CRITICAL", it can't get worse.
        if 'CRITICAL' in states:
            continue

        # 3.a Evaluate list of disk status flag.
        if disk_status in ["Normal"]:
            states.append('OK')
        elif disk_status in ["SystemPartitionFailed", "Crashed"]:
            states.append('CRITICAL')

        # 3.b Evaluate disk temperature thresholds.
        if warning and warning < int(disk_temp):
            states.append('WARNING')
        if critical and critical < int(disk_temp):
            states.append('CRITICAL')

        # 3.c Evaluate list of disk health status flag.
        if disk_health_status in ["Normal"]:
            states.append('OK')
        elif disk_health_status in ["Warning"]:
            states.append('WARNING')
        elif disk_health_status in ["Critical", "Failing"]:
            states.append('CRITICAL')

    # 4. Compute outcome for overall sensor state.
    state = 'UNKNOWN'
    priorities = ['CRITICAL', 'WARNING', 'UNKNOWN', 'OK']
    for priority in priorities:
        if priority in states:
            state = priority
            break

    # 5. Respond with textual and perfdata output and propagate exit code.
    print('%s%s %s' % (state, output, perfdata))
    exitCode()

if mode == 'storage':
    output = ''
    perfdata = '|'
    for item in snmpwalk('1.3.6.1.2.1.25.2.3.1.3'):
        i = item.oid_index or item.oid.split('.')[-1]
        storage_name = item.value
        if re.match("/volume(?!.+/@docker.*)", storage_name):
            allocation_units = snmpget('1.3.6.1.2.1.25.2.3.1.4.' + str(i))
            size = snmpget('1.3.6.1.2.1.25.2.3.1.5.' + str(i))
            used = snmpget('1.3.6.1.2.1.25.2.3.1.6.' + str(i))

            storage_size = int((int(allocation_units) * int(size)) / 1000000000)
            storage_used = int((int(used) * int(allocation_units)) / 1000000000)
            storage_free = int(storage_size - storage_used)

            # some virtual volume have size zero
            if storage_size == 0:
                continue

            storage_used_percent = int(storage_used * 100 / storage_size)

            if warning and warning < int(storage_used_percent):
                if state != 'CRITICAL':
                    state = 'WARNING'
            if critical and critical < int(storage_used_percent):
                state = 'CRITICAL'

            output += ' -  free space: ' + storage_name + ' ' + str(storage_free) + ' GB (' + str(storage_used) + ' GB of ' + str(storage_size) + ' GB used, ' + str(storage_used_percent) + '%)'
            perfdata += storage_name + '=' + str(storage_used) + 'c '
    print('%s%s %s' % (state, output, perfdata))
    exitCode()

if mode == 'update':
    update_status_nr = snmpget('1.3.6.1.4.1.6574.1.5.4.0')
    update_dsm_version = snmpget('1.3.6.1.4.1.6574.1.5.3.0')
    status_translation = {
            '1': "Available",
            '2': "Unavailable",
            '3': "Connecting",
            '4': "Disconnected",
            '5': "Others"
        }
    state_translation = {
        '2': 'OK',
        '1': 'WARNING',
    }

    update_status = status_translation.get(update_status_nr)
    state = state_translation.get(update_status_nr, "UNKNOWN")

    print(state + ' - DSM Version: %s, DSM Update: %s' % (update_dsm_version, update_status), '| DSMupdate=%sc' % update_status_nr)
    exitCode()

if mode == 'status':

    # 1. Retrieve and decode system metrics.
    status_model = snmpget('1.3.6.1.4.1.6574.1.5.1.0')
    status_serial = snmpget('1.3.6.1.4.1.6574.1.5.2.0')
    status_temperature = snmpget('1.3.6.1.4.1.6574.1.2.0')

    status_system_nr = snmpget('1.3.6.1.4.1.6574.1.1.0')
    status_system_fan_nr = snmpget('1.3.6.1.4.1.6574.1.4.1.0')
    status_cpu_fan_nr = snmpget('1.3.6.1.4.1.6574.1.4.2.0')
    status_power_nr = snmpget('1.3.6.1.4.1.6574.1.3.0')

    status_translation = {
        '1': "Normal",
        '2': "Failed"
    }

    status_system = status_translation.get(status_system_nr)
    status_system_fan = status_translation.get(status_system_fan_nr)
    status_cpu_fan = status_translation.get(status_cpu_fan_nr)
    status_power = status_translation.get(status_power_nr)

    # 2. Compute outcome for overall sensor state.

    # 2.a Evaluate list of system status flags.
    status_all = [status_system, status_system_fan, status_cpu_fan, status_power]
    if all([status == "Normal" for status in status_all]):
        state = 'OK'
    elif any([status == "Failed" for status in status_all]):
        state = 'CRITICAL'
    else:
        state = 'UNKNOWN'

    # 2.b Evaluate system temperature thresholds.
    # When state is already "CRITICAL", it can't get worse.
    if state != 'CRITICAL':
        if warning and warning < int(status_temperature):
            state = 'WARNING'
        if critical and critical < int(status_temperature):
            state = 'CRITICAL'

    # 3. Respond with textual and perfdata output and propagate exit code.
    print(state + ' - Model: %s, S/N: %s, System Temperature: %s C, System Status: %s, System Fan: %s, CPU Fan: %s, Powersupply : %s' % (status_model, status_serial, status_temperature, status_system, status_system_fan, status_cpu_fan, status_power) + ' | system_temp=%sc' % status_temperature)
    exitCode()