Monitoring Intel/LSI/Symbios RAID controllers

[artem.ultra]

now

PS C:\zabbix> .\raid_discovery.ps1
info from server: "processed: 4; failed: 0; total: 4; seconds spent: 1.154692"
sent: 4; skipped: 0; total: 4

PS C:\raid> .\CmdTool2_64.exe -adpallinfo -a0
User specified controller is not present.
Failed to get CpController object.

Exit Code: 0x01

- hw.raid.discovery.pdisks { "data":[{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"8", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"7", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"1", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"10", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"0", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"2", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"5", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"3", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"4", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"6", "{#ADAPTER_ID}":"0" },{ "{#ENCLOSURE_ID}":"16", "{#PDRIVE_ID}":"9", "{#ADAPTER_ID}":"0" } ]}
- hw.raid.discovery.vdisks { "data":[{ "{#VDRIVE_ID}":"1", "{#ADAPTER_ID}":"0" },{ "{#VDRIVE_ID}":"0", "{#ADAPTER_ID}":"0" },{ "{#VDRIVE_ID}":"2", "{#ADAPTER_ID}":"0" } ]}
- hw.raid.discovery.bbu { "data":[{ "{#ADAPTER_ID}":"0" } ]}
- hw.raid.discovery.adapters { "data":[{ "{#ADAPTER_ID}":"0" } ]}

[artem.ultra]

[ArtemK]

well, everything looks ok, now you need to execute raid_trapper_checks.ps1 script on your server, it will send metrics about your raid controller to zabbix server. Use windows scheduler to execute it automatically for example every 5 minutes.

[artem.ultra]

It proved necessary to update perc cli

[Sugarman]

Hey guys

After some fiddling around with paths I got the scripts running nicely. Thank you for all the work, ArtemK.

Has anybody figured out a similar way to monitor Adaptec controllers?

[Sugarman]

Just to answer my own question:

https://github.com/lesovsky/zabbix-extensions/tree/master/files/hwraid-adaptec

This works roughly the same for Adaptec controllers as Artems scripts. Lesovskys do need a few syntax updates to have them working with V3.0 and debian paths.

[VladimirM]

ThinkServer

Does anybody know template for Lenovo ThinkServer (ThinkServer RAID, smart ...) ?

[philipnet]

Hi,

Just to say that this template and scripts really helped setting up RAID monitoring on two new servers.
Thanks.

I had to make a couple of alterations in order to fit our environment:
1) Add " on {HOST.NAME}" to the triggers so that we knew which host was alerting
2) Add " -NoLog" to the MegaCli64 command so that MegaCli64 doesn't add/create a log file

[ArtemK]

Glad it worked for you!

2) Add " -NoLog" to the MegaCli64 command so that MegaCli64 doesn't add/create a log file

there is "-nolog" switch in all commands. You probably used "old" scripts from archive attached to the first post in this topic. I've replaced it now with link to github repo.

[philipnet]

I downloaded the ones from http://www.zabbix.org/wiki/Templates/Intel_LSI_RAID

I'm unable to create an account on the Wiki, so I can't update that page


(I get the error "Account creation error, There was either an authentication database error or you are not allowed to update your external account.", which I'll take up elsewhere )

[ArtemK]

ah, I completely forgot that there is article in wiki. Thanks for the link, I'll try to update that article soon and I'll also post this template on share.zabbix.com.

[iqarmy]

Install help

Hi guys,

I'm trying to get this template installed and working within my zabbix environment.

Can someone help me out with the basics?

I've done this thus far:

1. Imported the template into the zabbix server
2. Associated the template with the host I intend to monitor
3. Deployed the scripts within the target machine which runs CentOS 7 thus my cli path is: my $cli = '/opt/MegaRAID/MegaCli/MegaCli64';

Now I don't know what to do next as the discovery scripts outpus this:

scripts]# ./raid_discovery.pl
info from server: "processed: 0; failed: 3; total: 3; seconds spent: 0.000293"
sent: 3; skipped: 0; total: 3

Whilst the trapper_check one writes the following into: /tmp/raid-discovery-zsend-trapper-data.tmp
..........
- intel.raid.physical_disk[0,4,13,"inquiry_data"] "NCG0DA6L HGST HUS726060ALE610 APGNT517"
- intel.raid.physical_disk[0,4,14,"media_errors"] "0"
- intel.raid.physical_disk[0,4,14,"predictive_errors"] "0"
- intel.raid.physical_disk[0,4,14,"raw_size"] "5.458 TB"
- intel.raid.physical_disk[0,4,14,"firmware_state"] "Online, Spun Up"
- intel.raid.physical_disk[0,4,14,"inquiry_data"] "NAHS71SX HGST HUS726060ALE610 APGNT7J0"


Please advise.

[ArtemK]

Hi!

..........
- intel.raid.physical_disk[0,4,13,"inquiry_data"] "NCG0DA6L HGST HUS726060ALE610 APGNT517"
- intel.raid.physical_disk[0,4,14,"media_errors"] "0"

I've used keys like %intel.raid% in previous version of these scripts. Maybe you have imported updated template, but "old" scripts in use?

try to use template and scripts from this github repo.

[iqarmy]

Hi ArtemK,

I'm using the intel version template + scripts with trapper

As mentioned can you walk me through the basics please

for instance running from cli the ./raid_check.pl script I get:

[root@storage3 scripts]# ./raid_check.pl
Mode is not defined. Use --mode paramater at ./raid_check.pl line 20.

Does the agent needs to be configured in active mode or can the server pull the data once associated the template with it and warn for offline drives broken VD's etc?

[iqarmy]

template import error

When trying to import the template from git I'm getting the following error:

Cannot find value map "LSI RAID BBU & LD Status" used for item prototype "BBU state on adapter:{#ADAPTER_ID}" of discovery rule "RAID discovery bbu" on "Template LSI RAID".