coolspot18
Contributor
Contributor

vSphere ESXi 6.5.0 Hardware Sensors

Jump to solution

Hello Everyone,

I have a Q170 based motherboard and have successfully loaded ESXi 6.5.0. However, ESXi does not display any Hardware Sensor Data and reports "No Sensor Data Available".

Does ESXi require BMC or IMPI to obtain hardware database? Does it not get at least basic sensor information from the CPU sensors, etc.?

Thanks

41 Replies
rmmckernan
Contributor
Contributor

I have add the following message to under the manager request option

Good Morning, It has been almost 3 weeks and the only thing that I have been told is that this is a bug and engineering team is working on it.  I have not been provided any proof as to when this will be resolved , the priority, or if there is any expectation that this will be resolved on the next patch release. Please , Please provide more details outside of "we are working on it"

Thanks,

0 Kudos
PROXSYS
Contributor
Contributor

Thank you for the update!

When this is taking to long i will also open a case. Please keep me updated

0 Kudos
RLI2013
Enthusiast
Enthusiast

The Workaround did not help for me. Even after removing the Service I still have "no host data available" under sensors. Is a hostd or vCenter Service restart necessary?

esxcli system wbem get shows the following:

   Authorization Model: password

   Enabled: true

   Loglevel: warning

   Port: 5989

   WSManagement Service: true

which should be ok.

What I recognized is after removing the Service as explained in KB2148520, if I do search again I'll find the Service "cis.vws.servicename" again but with a different new ServiceID. Is this expected?

Thanks

0 Kudos
rmmckernan
Contributor
Contributor

Hey Everyone,

I still have an outstanding case open with VMware, They have identified it as and issue and repeated the installation on software on there side. Based on what I am being told they are have filed a report with engineering team.

We don't have an update as of today. The case has been open for 6 weeks.

At this time I don't think it is wise to try any of the fixes listed above on production equipment. if you have test equipment ; full speed ahead. But at this time VMware does not have a fix.

As a side not, Please make sure the following service is started on the ESX host. Here is a link to the VMware information regarding starting this service

The Hardware Status tab displays the error: Hardware monitoring service on this host is not respondi...

If you have esx 6.5 installed you can review services from the host Webclient

Host>Manage>Services>sfcbd-watchdog>highlight and click start.

From Vcenter webclient

Select Host and Clusters > Select Host > Config> Security Profile > Services > EDIT > CIM Service > Start

0 Kudos
LittleCrow
Contributor
Contributor

I have this exact same issue on our DL380G8s, and have been following this thread with interest.

Some new VIB's were released on the 11th July, I've pushed these out to one of our servers with Update Manager, but this didn't resolve the issue.

This morning I checked the HPE vibsdepot and it looks like new vibs are in the pipeline.  There is a folder that is very intermittently available @ http://vibsdepot.hpe.com/hpe/jul.18.2017/

The intermittent behaviour happened with the earlier July release too and eventually went away, so I assume it's just being replicated slowly across their CDN.

Fingers crossed this one contains the fix!

0 Kudos
bphinz
Contributor
Contributor

It doesn't appear that the July 18 vibs resolved the problem.  I still have no sensors showing up on any of my hosts

0 Kudos
LittleCrow
Contributor
Contributor

I've updated one of our boxes to the just released 6.5U1 using the custom HPE image, and the same issue is present.

Our issue is the same as posted by PROXSYS, we get some Hardware sensors just not as many as we saw in ESXi 5.5 - The temperature sensors being the main stand out.

It also looks like they've removed the ability to connect direct to the host using the 6.0 'full fat' client in U1, but checking another 6.5 box the sensors aren't populated there either.

0 Kudos
rmmckernan
Contributor
Contributor

Just wanted to update this thread.

Over the weekend a VMware engineer updated my case. .... Yes it is still open.

Please have a look at the following links ..... There is no resolution at this time but they have confirmed the bug and are working on it.

https://kb.vmware.com/kb/2151238

https://recommender.vmware.com/solution/SOL-21564

Thanks,

Cherubim
Contributor
Contributor

Hi,

do you have any news on this topic? I've got the same issue with HPE BL460c G9 since I updated to ESXi 6.5 U1

0 Kudos
rmmckernan
Contributor
Contributor

Hey Cherubim ,

Please see the link in the post above , and Monitor the Resolution section. They do not have an update/fix at this time.

I have a  around to the issue by pulling SNMP counters about hardware health from HP iLO . You may be able to do the same for Dell iDrac.

HP also has API ability on the ILO that you may be able to pull hardware status from as well. See link below

https://support.hpe.com/hpsc/doc/public/display?docId=mmr_kc-0126906

Thanks,

Ryan

0 Kudos
hmaATrms
Contributor
Contributor
0 Kudos
Darkpacific
Contributor
Contributor

Good day !

There is a problem displaying the sensors in VCSA 6.5 received from the host esxi 5.5, VCSA 5.5 information sensors 208 is shown in detail sensors and VCSA 6.5 only 147 sensors ! esxi 5.5 installed on the HP ProLiant DL360 G5, anyone has this problem ?

5.5 in vcsa 5.5.jpg5.5 in vcsa 6.5.jpg

0 Kudos
r-anacker
Contributor
Contributor

Hello i have the same Problem:

VCSA 6.5 won't display the HP DL380 G5 storage status the esxi Client 5.5 display the storage status.

Did you open a support case?

0 Kudos
rmmckernan
Contributor
Contributor

The case has been closed but a KB article was started

https://kb.vmware.com/kb/2151238

https://recommender.vmware.com/solution/SOL-21564

I have also posted another solution using HP iLO to pool SNMP for hardware information. We are currently using PRTG to monitor . PRTG has a build template for HP iLO SNMP.

There is a free version for 100 Sensors depending on the size of your environment.

You may also be able to query HP iLO API for information , I post a link on a couple previous reply's.

you can PM me if you would like.

Thanks,

Ryan

0 Kudos
mvrk
Enthusiast
Enthusiast

Hi,

I was trying the workarounds, but for me none work.

The service (cis.vws.servicename) doesn't exist on the mob search.

And the other one on esxi (wbem) is enabled.

Btw, according to kb https://kb.vmware.com/s/article/2148520​ this problem was resolved on 6.5 U1d, but i used the newer 6.5 U1e to upgrade from 6.0 U2 and the problem still here, maybe somehow they introduced the problem again on U1e. I have not yet tried to upgrade using 6.5 U1d to see if the problem happens or not...

0 Kudos
mvrk
Enthusiast
Enthusiast

Just updated using 6.5 U1d and still same problem.

0 Kudos
daescobar
Contributor
Contributor

Hi!

We are experiencing the same problem when we upgrade from ESXi 5.5 to 6.5. Our hosts are HP Blade BL460c. We see 163 sensors in ESXi 5.5 but just only 51 in 6.5 Smiley Sad

We wonder if someone has found a solution to this issue.

Thank you very much!! Any help will be appreciated.

0 Kudos
jsprinkleISG
Enthusiast
Enthusiast

As indicated in this https://kb.vmware.com/kb/2151238 article, they've purposely removed every sensor there that doesn't report a health state. They have moved or will be moving that data to elsewhere in the UI.

daescobar
Contributor
Contributor

Thank you very much jsprinkleISG for your response.

You are right, we can see the sensors if we open the UI of the host (not via vCenter). But we see only a few of them (55):

pastedImage_0.png

Nevertheless, in the previous version we can see much more sensors (163).

pastedImage_2.png

Regards,

0 Kudos
IBFsupport
Contributor
Contributor

Hi,

We had the same issue on 2 hosts vmware esxi 5.5 and 6.0

The only solution I have got to work so far is:

1. Connect to the vmware esxi host by SSH.

Login with root

2. localcli hardware ipmi sel clear

3. /etc/init.d/sfcbd-watchdog restart

4. services.sh restart

You can check the status on the vcenter to see if the host have nbow change from red/yellow to green

https://your_vcenter_server_name_here_FQDN/mob

And then follow this beloe path for verifying that the health now have changed from red/yellow to green.

content -> rootFolder -> childEntity-> HostFolder -> (childEntity) -> Host -> runtime -> healthSystemRuntime -> hardwareStatusInfo -> memoryStatusInfo

0 Kudos