ESXi 6.7 monitor hardware storage not showing usual results

Question

Yesterday 2 of our ESXi hosts were updated from 6.5 to 6.7 (build 8169922), though after doing this we noticed a few quirks when trying to access the storage hardware monitoring.

In the web UI typically we'd only see elements listed for the Avago/LSI hardware RAID here, with a clear indicator of its overall health, each logical volume's health, and each physical disk's health. Now though all sorts of hardware is showing up that isn't even storage related. (Such as the GPU, Ethernet controllers, and Thunderbolt controllers.)

I've updated the LSI SIMS provider to version 7.6 (30/May/2018), along with MegaCLI utility version 8-07-07, and the storcli utility version 007.0606.0000.0000 (20/Mar/2018). The host has been rebooted after installing each VIB. Looks like the CIM service just doesn't want to know though.

The servers work very fine indeed, VMs are operating as normal. They just lack RAID monitoring and management now, so if an issue happens we just get a buzzer and LED blink, and have to reboot to the LSI WebBIOS to see what it's for to resolve it.

I saw from the VMWare hardware compatibility that our LSI RAID controller models are supported with an official driver so I don't think there is any compatibility issue.

The MegaCLI command is also useless likely because of the same CIM issue. ./MegaCli -adpCount shows Controller Count: 0, which also means that the MegaRAID Storage Manager utility no longer detects ESXi hosts for management too.

--

Finally with the storcli only some parts of it work. This is probably a separate issue though. (Bug in the tool?) It seems if you are specifying command arguments that contain a forward slash '/' it is ignored.

For example, "storcli show" works:

[root@nexus:/opt/lsi/storcli] ./storcli show
CLI Version = 007.0606.0000.0000 Mar 20, 2018
Operating system = VMkernel 6.7.0
Status Code = 0
Status = Success
Description = None

Number of Controllers = 1
Host Name = nexus.***
Operating System  = VMkernel 6.7.0
StoreLib IT Version = 07.0603.0200.0000
StoreLib IR3 Version = 15.53-0

System Overview :
===============

------------------------------------------------------------------------------------
Ctl Model                 Ports PDs DGs DNOpt VDs VNOpt BBU  sPR DS  EHS ASOs Hlth
------------------------------------------------------------------------------------
  0 LSIMegaRAIDSAS9271-8i     8   4   1     1   1     1 Msng On  1&2 Y      4 NdAtn
------------------------------------------------------------------------------------

Ctl=Controller Index|DGs=Drive groups|VDs=Virtual drives|Fld=Failed
PDs=Physical drives|DNOpt=DG NotOptimal|VNOpt=VD NotOptimal|Opt=Optimal
Msng=Missing|Dgd=Degraded|NdAtn=Need Attention|Unkwn=Unknown
sPR=Scheduled Patrol Read|DS=DimmerSwitch|EHS=Emergency Hot Spare
Y=Yes|N=No|ASOs=Advanced Software Options|BBU=Battery backup unit
Hlth=Health|Safe=Safe-mode boot

However, using another command such as "storcli /cx show alarm" does not:

[root@nexus:/opt/lsi/storcli] ./storcli /cx show alarm

syntax error, unexpected $end

     Storage Command Line Tool  Ver 007.0606.0000.0000 Mar 20, 2018

     (c)Copyright 2018, AVAGO Technologies, All Rights Reserved.


help - lists all the commands with their usage. E.g. storcli help
<command> help - gives details about a particular command. E.g. storcli add help

List of commands:

That syntax error, unexpected $end suggests something in the command parser is not right. (I've tried from both the DCUI and SSH.) Weirdly storcli cx show help does work, only showing controller specific commands, but storcli cx show all does not.

(The fact that storcli is able to see the controller at all and query it tells me the driver is fine and compatible.)

Solution 1

You should use

./storcli /c0 show

(0 = number of the controler, you see it in your first command ./storcli show -> Ctl 0)

If you want to see every controller you can use :

./storcli /call show

If you want to see every virtual drive you can use :

./storcli /call/eall show

If you want to see every physical drive you can use :

./storcli /call/eall/sall show

Or you can use :

./storcli /call show all 

for all the information of all controllers.

In your case you have 1 controller, so you can use

./storcli /c0 show all

For alarm status/control you can use :

./storcli /c0 show alarm   (show alarm status)

./storcli /c0 set alarm=silence   (mute buzer alarm)

./storcli /c0 set alarm=off   (shutdown alarm)

Solution 2

Based on the list https://www.vmware.com/resources/compatibility/pdf/vi_cim_guide.pdf last version where LSIProvider mib is support was 6.5u2

Comments

[Adambean]

Thanks, I got this working already. (Comment 21 Jun at 14:55) This won't fix the ESXi UI though.

[IceCubee-FR]

Yes same issue here with a 9260-8i on ESXi 6.7. I use a script for check the raid status with nagios but you can use directly on a linux machine with an ssh acces to the ESXi.