did some research on how to get similar results when using the local plugin “smart” on linux systems under Windows. Wasn’t able to find a script. So I’ve created a PowerShell (my first one , so please excuse ugly or non-optimal code) script.
Testing and adopting still ongoing, but at least for a standard dedicated windows server it delivers the information required and integrates immediately within Check-mk doing an inventory.
Would be happy to get some feedback.
BTW - PowerShell has got some really nice features, especially the JSON management impresses me.
Requires Smartmontools being installed.
Code is being maintained on github, planning to deliver a package some time.
I was searching for this a lot. Does your script works for SSD wear indication as well?
If not I’m working on the script that will cover all possibilities, but in vbs. If u care to share your knowlege contact me so we could make something for the community
in the end the input for the script is the output of smartctl. If smartctl is delivering the SSD wear information, it could be parsed.
Haven’t worked on the script for a while, as most of my windows servers are now virtualized, so the disk monitoring is done on another point.
But - in the end - is a good idea to do this on the ESX level as well.
The value you are looking for is 177 - Wear_Leveling_Count
If your SSD is delivering the value, it will be visible by my script:
In this case - it is just a cheap one - the smart command isn’t providing wear_level_count.
Please test on your system:
a) get a list of all disks attached —> smartctl --scan
b) get the current values —> smartctl -a -d /dev/sda (if this is a name in your list)
If you can see here the wear level, all is good. If not - it is not provided by the SSD.
I have tested the script and everything seems to be working. I just need to check reporting for nvme. I read something about nvme.cli but only for some nvme’s. Your sccript is what I was looking for in a week now. Thanks a lot!
Best regards,
Hi,
the output is truncated, not sure, what the issue is about.
As already posted - please check first the output of smartclt, the script isn’t that complicated, parses the smartctl output and delivers in a format, which can be analyzed by check_mk.