Hello,
We use checkmk to monitor over redfish Plugin our XCCs from the SR650v3 servers.We recently added two new servers, each with a RAID B540p-2HS RAID controller.
In XCC, everything is displayed as green, and there are no errors or events.
However, our monitoring, which queries XCC via Redfish, shows the following error:
Type: SAS3808N, RaidLevels: RAID0,RAID1,RAID10,RAID00, DeviceProtocols: SATA,SAS,NVMe
Details
Component State: Normal, This resource is disabled.
What could this mean?
CheckMK Version : 2.3.0p40
Redfish Plugin Version : 2.3.77 from Andreas .
Thanks
Michael
This means normally that this Raid controller is not used (no drives connected or anything like that).
But more can only be said if i can have a look at the agent section for the raid controllers.
Hello Andi ,
What u exactly need from me ?
Thanks .
This is what Lenovo wrote :
Your description indicates: "In XCC, everything is displayed in green, and there are no errors or events. However, our monitoring system reports the following error via Redfish querying XCC:
Type: SAS3808N, RAID Levels: RAID0, RAID1, RAID10, RAID00, Device Protocols: SATA, SAS, NVMe
Details
Component Health: Normal, This resource is disabled."
Since XCC displays green and no events are present, it can be assumed that the volumes still exist and the controller was initialized normally. It is very likely that this is an interpretation issue with the Redfish report and not a hardware or configuration problem.
“Component Health: Normal” → means that the hardware is healthy.
“Resource is disabled” → means that the Redfish resource endpoint is not active for management operations.
Your B540p-2HS RAID controller is healthy and functioning normally.
That is simply wrong 
The definition says –> “Disabled - This function or resource is disabled.”
Nothing else. But as i said before - only the raw agent output for this device can give us more insight if there is a real problem or a firmware bug present.
Hello Andi , again the Question , how can i create an Agent Output from an Redfish Service / Host ? Thanks
Michael
Sorry if i missed this before. Normally you can click on the Check_MK service hamburger menu and select “Download agent output”.
As i had some time now over the holidays i had a look at the agent output and here are my findings 
It looks like the SAS3808N, as it is not a real controller, has some type of a firmware bug that is reporting the false state. Not the health state only the operational state.
The second agent output you sent has a real raid controller (5350-8i) and don’t show this i hope false state message.
compared with
What i can do for such controllers is only to implement a “Raid Storage Rollup” service. Like it is also done for the memory summary check.
The “HealthRollup” state of you two systems are completely fine on booth systems.
Would this approach a viable option in your case?
1 Like
Hello ,
Sorry was three weeks not in the Office .
Yes this will be a very good Option for me and many thanks for your effort .
Means this you implement this in a new redfish Package ?
Michael
First i will include this inside the Redfish extension mkp that was posted yesterday.
If it is working on the monitored systems without problem i will make this then also as a PR on github for the mainline CMK repo.
Redfish extensions for existing integration
1 Like
