ScaleFlux CSD monitoring

checks-sfx_csd_status.py

#!/usr/bin/python
# -*- encoding: utf-8; py-indent-offset: 4 -*-


# alert for erroneous status from ScaleFlux CSD to Check_MK for monitoring
# capacity / temp etc.
# part of my benchmarks / best practice
# complete hack job because i'm tired

#SFX card: /dev/sfdv0n1
#PCIe Vendor ID:                    0xcc53
#PCIe Subsystem Vendor ID:          0xcc53
#Manufacturer:                      ScaleFlux
#Model:                             CSD 2000 Series
#Serial Number:                     UC1945A7118M
#OPN:                               CSDU3RF040B1
#FPGA BitStream:                    4705
#Drive Type:                        U.2-V
#Software Revision:                 3.1.1.0-51107
#Temperature:                       54 C
#Power Consumption:                 11 W
#Atomic Write mode:                 OFF
#Percentage Used:                   0%
#Data Read:                         2389 GiB
#Data Written:                      2739 GiB
#Correctable Error Cnt:             0
#Uncorrectable Error Cnt:           0
#Check Log:                         0
#PCIe Link Status:                  Gen3 x4
#PCIe Device Status:                Good
#Disk Capacity:                     3840 GB
#Physical Capacity:                 3840 GB
#Compression Ratio:                 120%
#Physical Used Ratio:               64%
#Critical Warning:                  0


# add parser for n cards >1
def parse_sfx_csd_status(info):
    parsed = {}
    for line in info:
        if line[0] == "SFX card":
            item = line[1].replace(" /dev/","")
            parsed[item] = {}
        else:
            parsed[item][line[0]] = line[1].lstrip()

    return parsed


def inventory_sfx_csd_status(info):
    parsed = parse_sfx_csd_status(info)
    #import pprint
    #pprint.pprint(parsed)
    for entry in parsed:
        yield entry, None


def check_sfx_csd_status(item, params, info):
    state = 0
    msg = []
    parsed = parse_sfx_csd_status(info)
    if   int(parsed[item]["Critical Warning"]) != 0:
        msg += [ "has Critical Warning" ]
        state = 2

    temp = int(parsed[item]["Temperature"].replace(" C",""))
    if   temp >= 60:
        msg += [ "is overheating (%d) C" % temp ]
        state = 2

    if   int(parsed[item]["Uncorrectable Error Cnt"]) > 0:
        msg += ["has uncorrected errors"]
        state = 2

    if   int(parsed[item]["Percentage Used"].replace("%","")) >= 99:
        msg += ["is full"]
        state = 2

    if   state != 0:
        yield state, (" ").join(msg)
    else:
        yield state, "is operating normally"



check_info['sfx_csd_status'] = {
    'inventory_function': inventory_sfx_csd_status,
    'check_function': check_sfx_csd_status,
    'service_description': 'CSD Status %s',
}

plugins-sfx_csd_status.sh

#! /usr/bin/env bash
  
# forward status from ScaleFlux CSD to Check_MK for monitoring
# capacity / temp etc.
# part of my benchmarks / best practice

#SFX card: /dev/sfdv0n1
#PCIe Vendor ID:                    0xcc53                             
#PCIe Subsystem Vendor ID:          0xcc53                             
#Manufacturer:                      ScaleFlux                          
#Model:                             CSD 2000 Series                    
#Serial Number:                     UC1945A7118M                       
#OPN:                               CSDU3RF040B1                       
#FPGA BitStream:                    4705                               
#Drive Type:                        U.2-V                              
#Software Revision:                 3.1.1.0-51107                      
#Temperature:                       54 C                               
#Power Consumption:                 11 W                               
#Atomic Write mode:                 OFF                                
#Percentage Used:                   0%                                 
#Data Read:                         2389 GiB                           
#Data Written:                      2739 GiB                           
#Correctable Error Cnt:             0                                  
#Uncorrectable Error Cnt:           0                                  
#Check Log:                         0                                  
#PCIe Link Status:                  Gen3 x4                            
#PCIe Device Status:                Good                               
#Disk Capacity:                     3840 GB                            
#Physical Capacity:                 3840 GB                            
#Compression Ratio:                 120%                               
#Physical Used Ratio:               64%                                
#Critical Warning:                  0    


if ! type sfx-status >/dev/null; then
    exit 0
fi


echo '<<<sfx_csd_status:sep(58)>>>'
sfx-status