vSphere ESXi 6.5.0 Hardware Sensors

[coolspot18]

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

[srwsol]

I'm having the same problem.  Originally I thought the problem was related to the new version of vCenter server in conjunction with ESXi 6.0u2, as I upgraded vCenter server first.  However, then I upgraded one of my hosts to 6.5 from 6.0u2 and after doing that I found that the sensors were still missing from the new version of vCenter server, and they are also missing from the html5 client on the upgraded host.   However, the sensors are still present if you connect directly to the upgraded host from the old C client (you can't connect the old C client to the new version of vCenter Server).

[motuzko]

Same situation. "No sensor data available" on Monitor/Hardware tab. Is there way to figure out why? Which log files should be checked?

[HelloBJ]

What is the output with "vsish -e get /hardware/ipmiSupport" ?

Thanks!

[motuzko]

[root@ESXi:~] vsish -e get /hardware/ipmiSupport

IPMI state: 0 -> IPMI not supported

Hardware

Shuttle Global - SH67H3

I thought I had hardware data available with ESXi 5.x/C client

[srwsol]

Yes, the sensor data does display perfectly if you connect the native C client directly to the ESXi 6.5 host, so this is not an issue of ESXi 6.5 being unable to obtain the sensor data from the hardware.   It does appear to be an issue with how that data is made available to the GUI interfaces, both the native html5 interface within ESXi 6.5 itself and the GUI interfaces within vCenter Server.   I've sinced rolled back ESXi 6.5 to 6.0u2 due to problems with host connected USB devices and VMs, so I can't say whether or not the sensor data was being presented correctly via the CLI interface as I didn't try that.

[andy213ca]

Hi Srwsol,

Could you please elaborate a bit on "native c client" directly able to successfully obtain CPU temperature data from ESXI host?

we are running into similar problem with ESXI 7.0 and ASUS desktop motherboard (no ipmi capability).  We would like to use any custom C code available to run on the cli to view the cpu temperatures  please. 

 

[RAJ_RAJ]

Hi ,

vSphere 6.5 is not supported most of the hardware , kindly please check the HCL for details .

http://www.vmware.com/resources/compatibility/search.php

[EstherP]

Hi everyone

I think is not a problem related to the hardware compatibility.

So far I have done the following tests:

vCenter 6.0U1b (windows) with sensors WORKING ---> upgrade to vCenter 6.5 ((windows) - Sensors NO working

vCenter 6.0U1b (windows) with sensors WORKING ---> migrate to vCenter 6.5 (vapp) - Sensors NO working

Clean installation vCenter appliance 6.5 --> sensor WORKING (with support hardwarde HP-G8, and with unsupport hardware HP G6, also working)

I am suffering the problem with the sensors in vCenter 6.0 (since the upgrade from 5.5 to 6.0) . At present, the sensors  are only visible, if I reset and update the Hardware sensors, but vCenter is unable to refresh the sensors automatically. HP and VMware confirmed to me, that they were able to reproduce the same problem in their Labs and the solution would be implemented in the next releases.

We are even worse, because in that case, the sensors are not visible anytime.

At the beginning, I thought it could be due to a lack of permission, but even with the [e-mail], are not visible. I have been comparing all the advance options between the fresh installation and the vcenter already migrated, and apparently, there are not differences.

I will open a new case to HP and VMware, and we will see if they can provide us any solution. Meanwhile, I will continue doing different tests in different scenarios.

If someone has more information, please let me know.

Regards

Piki

[Hedgehog_57]

little investigation.

VCSA 6.5

Entering by SSH.

There are two interesting logs

/var/log/vmware/vsphere-ui/logs/vsphere_client_virgo.log

This is HTML5 client log. When accessing any of Hardware Health you get this in the logfile:

-----------------------------------------------------

[2016-12-28T14:37:33.929Z] [ERROR] data-service-pool-1112       70000225 100003 200001 com.vmware.vise.util.http.ConnectionManager                       Error reading from the stream: https://vcenter.fqdn:443/hwh/gwtcim?host=host-5170&op=getHostData

[2016-12-28T14:37:33.929Z] [WARN ] data-service-pool-1112       70000225 100003 200001 com.vmware.vim.cimmonitor.http.impl.CimVc61HttpService            No host data. Http status code: 404

[2016-12-28T14:37:33.936Z] [ERROR] data-service-pool-1105       70000225 100003 200001 System.err                                                        [Fatal Error] :1:1: Premature end of file.

[2016-12-28T14:37:33.937Z] [ERROR] data-service-pool-1105       70000225 100003 200001 c.v.vsphere.client.host.hardware.ciminfo.HostCimInfoXmlReader     Error parsing XML document org.xml.sax.SAXParseException: Premature end of file.

        at com.sun.org.apache.xerces.internal.parsers.DOMParser.parse(DOMParser.java:257)

        at com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderImpl.parse(DocumentBuilderImpl.java:339)

        at javax.xml.parsers.DocumentBuilder.parse(DocumentBuilder.java:121)

        at com.vmware.vsphere.client.host.hardware.ciminfo.HostCimInfoXmlReader.initializeXmlDocumentAndXpath(HostCimInfoXmlReader.java:96)

        at com.vmware.vsphere.client.host.hardware.ciminfo.HostCimInfoXmlReader.parse(HostCimInfoXmlReader.java:81)

        at com.vmware.vsphere.client.host.hardware.ciminfo.HostCimInfoPropertyProvider.getHardwareStatus(HostCimInfoPropertyProvider.java:53)

        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)

        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)

        at java.lang.reflect.Method.invoke(Method.java:498)

        at com.vmware.vise.data.query.impl.ServicePropertyProviderAdapter.invokeMethod(ServicePropertyProviderAdapter.java:285)

        at com.vmware.vise.data.query.impl.ServicePropertyProviderAdapter.getProperties(ServicePropertyProviderAdapter.java:127)

        at com.vmware.vise.data.query.impl.DataManager.getDataFromPropertyProvider(DataManager.java:1301)

        at com.vmware.vise.data.query.impl.DataManager.getResultFromPropertyProvider(DataManager.java:1269)

        at com.vmware.vise.data.query.impl.DataManager.access$000(DataManager.java:78)

        at com.vmware.vise.data.query.impl.DataManager$1.call(DataManager.java:784)

        at com.vmware.vise.data.query.impl.DataManager$1.call(DataManager.java:780)

        at com.vmware.vise.util.concurrent.ExecutorUtil$2.call(ExecutorUtil.java:654)

        at com.vmware.vise.util.concurrent.ExecutorUtil$ThreadContextPropagatingTask.call(ExecutorUtil.java:957)

        at com.vmware.vise.data.query.internal.profiling.MdcAwareExecutorService$1.call(MdcAwareExecutorService.java:148)

        at java.util.concurrent.FutureTask.run(FutureTask.java:266)

        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)

        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)

        at java.lang.Thread.run(Thread.java:745)

-----------------------------------------------------

/var/log/vmware/vsphere-client/logs/vsphere_client_virgo.log

This is Flash client log. There are no exception but similar ERROR:

-----------------------------------------------------

[2016-12-28T15:04:40.152Z] [ERROR] data-service-pool-120259     70032048 102281 201989 com.vmware.vise.util.http.ConnectionManager                       Error reading from the stream: https://vcenter.technoserv.ts:443/hwh/gwtcim?host=host-5105&op=getHostData

[2016-12-28T15:04:40.153Z] [WARN ] data-service-pool-120259     70032048 102281 201989 com.vmware.vim.cimmonitor.http.impl.CimVc61HttpService            No host data. Http status code: 404

-----------------------------------------------------

Question is: what service is responsible for /hwh and where to see it's logs. Maybe it will give us a qlue.

[EstherP]

Hi again.

For those who upgraded vCenter from versión 6.0 to versión 6.5 and the "hardware sensors" were missing, with "no data available", here the First "workaround"

I have to say, that this workaround is not working for me completly, becuase after implementing, I can see all the hardware sensors visible (this is the part that is working), but if I remove a disk, or unplug a power cable, vCenter does not realize that there has been a hardware failure, and the warning is not created, unless I click manually in "Update"

I have done this in my testlab and it would be perfect if someone with the same problema, can run the workaround, and do some test, in order to see if the vCenter can automatically update, the hardware status.

[Hedgehog_57]

Maybe there is a way to reregister a service after removing?

[Hedgehog_57]

Can confirm. After removing this service We can see Hardware status of hardware. I don't have test lab at the moment to test auto update of removing some parts.

[RLI2013]

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

[PhyreMe]

SSH into your system logging in as ‘root’ with your administrator password.

You can check the status with ‘esxcli system wbem get’ and you’ll notice that enabled is ‘false’ by default.

Enable wbem by using esxcli system wbem set -e 1. Wait for a couple of minutes [I found it took about 15 before it showed], and then update the Hardware Health Status tab.

https://topics-cdn.dell.com/pdf/vmware-esxi-6.5.x_Release%20Notes_en-us.pdf

I also did this which helped on a host where the first did not

"No host data available" error in the Hardware Status tab after upgrading to vCenter Server 6.5 (2148520) | VMware KB

[JamieGator32]

"No host data available" error in the Hardware Status tab after upgrading to vCenter Server 6.5 (2148520) | VMware KB

This worked in my environment as well.  All seems to be functioning well now.  No more continuous 'refresh hardware information' messages from VEEAM.

This was driving me crazy since our upgrade on Saturday.  Thanks @phyrme for posting!

[Sergio Santos Jr]

This also seemed to somewhat correct the problem for me. I see sensor data now, but hardware events don't appear to trigger a sensor change. I'm going to do some more testing on this to verify.

Also noted the following:

1. 2 of the 3 PowerEdge R910 servers show the Serial number shows up as: "null" & the Tag field shows up with a super long GUID looking alpha-numeric string--there is no Asset Tag field. The 3rd PE R910 does show the Serial Number field with the proper DELL Service Tag, Tag field still shows the GUID number, and the Asset Tag field says "unknown". Due to them all being R910s I'm suspect the disparity is because of a differing firmware of the Lifecycle controller or other subsystem.

2. Both our PowerEdge R820s show the Serial Number field with the proper DELL Service Tag, Tag field shows the GUID number, and the Asset Tag field says "unknown".

3. Our PowerEdge R830 show the Serial number as: "null" & the Tag field shows up with the GUID looking alpha-numeric string--there is no Asset Tag field.

[FrancWest]

We are running HP Proliant DL360 gen 8 and gen 9. After upgrading, the sensor work, but I'm seeing way less sensors than with 6.0. For example, in 6.0 you saw the individual memory banks. Now with 6.5 it only display total memory, not individual banks. Also, some sensors in the tree even don't expand when you click on them.

[gerdesj]

"Also, some sensors in the tree even don't expand when you click on them"

Click on the expand all link in the list that starts "Update Reset Sensors ...", it looks like a little black triangle with four lines next to it.

[rmmckernan]

We also have a HP DL360 GEN 9 running ESXi 6.5. Same issue with sensors . I have opened a case with VMware and HP. HP claims that since they can see the hardware properly through iLO it is an OS issue.

VMware is still reviewing the issue. They have pulled system logs from my system. and are supposed to get back with me by next Monday.

At this point that have acknowledge that it is something "new" but not sure if it is a bug or just something specific with my system

Things I have tested, still on showing 79 sensors

• HP 6.5  customized image Nov2017 Install
• HP 6.5  customized image May2017 Install
• Latest 6.5 Image install , with Feb2017 Offline Bundle from HP , Including util bundle, and AMS bundling
• HP 6.0  U3 customized image Feb2017 Install ( 200+ sensors showed up)

Current build in production environment

• HP DL360 GEN 9, HPE-ESXi-6.0.0-Update2-iso-600.9.5.0.48 , VMware ESXi, 6.0.0, 3620759 ( 526 Sensors ) Showing.

Name                           Version                                Vendor           Acceptance Level  Install Date

-----------------------------  -------------------------------------  ---------------  ----------------  ------------

amshelper                      600.10.5.0-12.2494585                  HPE              PartnerSupported  2017-03-13

hpe-build                      600.9.4.5.11-2494585                   HPE              PartnerSupported  2016-02-24

hpe-cru                        6.0.8.6-1OEM.600.0.0.2494585           HPE              PartnerSupported  2017-03-13

hpe-esxi-fc-enablement         600.2.6.16-2494585                     HPE              PartnerSupported  2017-03-13

hpe-ilo                        600.10.0.0.26-1OEM.600.0.0.2494585     HPE              PartnerSupported  2017-03-13

hpe-smx-provider               600.03.11.00.9-2768847                 HPE              VMwareAccepted    2017-03-13

hp-conrep                      6.0.0.1-0.0.13.2159203                 Hewlett-Packard  PartnerSupported  2016-02-24

hpbootcfg                      6.0.0.02-01.00.11.2159203              Hewlett-Packard  PartnerSupported  2016-02-24

hpnmi                          600.2.3.14-2159203                     Hewlett-Packard  PartnerSupported  2016-02-24

hponcfg                        6.0.0.04-00.13.17.2159203              Hewlett-Packard  PartnerSupported  2016-02-24

hpssacli                       2.30.6.0-6.0.0.2159203                 Hewlett-Packard  PartnerSupported  2016-02-24

hptestevent                    6.0.0.01-00.00.8.2159203               Hewlett-Packard  PartnerSupported  2016-02-24

scsi-hpdsa                     5.5.0.46-1OEM.550.0.0.1331820          Hewlett-Packard  PartnerSupported  2016-02-24

scsi-hpsa                      6.0.0.114-1OEM.600.0.0.2494585         Hewlett-Packard  VMwareCertified   2016-02-24

scsi-hpvsa                     5.5.0.100-1OEM.550.0.0.1331820         Hewlett-Packard  PartnerSupported  2016-02-24

[PROXSYS]

We also experiencing this issue on HP DL360 GEN8 and GEN9 running ESXi 6.5 HP 6.5  customized image May 2017

Before the update (running HPE-ESXi-6.0.0-Update2) there are 192 sensors visible. After the update only 58 sensors are visible (see screenshots).

Did you get a response from VMWare?

[rmmckernan]

No update at this time, They have identified it as a bug and I have been told they are working on it.

I have not received any information on whether there will be a fix prior to the next update or not. I was told it is a bug with the CIM service

I will press for an update today. I am mildly annoyed that they don't have a place to list the bugs they are working on .

Thanks,

[rmmckernan]

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,

[PROXSYS]

Thank you for the update!

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

[hmaATrms]

""No host data available" error in the Hardware Status tab after upgrading to vCenter Server 6.5 (2148520) | VMware KB"

This helped me too. Thank you very much @PhyreMe.

[rmmckernan]

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 responding or not available …

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

[rmmckernan]

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]

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

[rmmckernan]

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

[Tom Croft]

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!

[bphinz]

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

[Tom Croft]

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.

[Darkpacific]

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 ?

[r-anacker]

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?

[rmmckernan]

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

[mvrk]

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...

[mvrk]

Just updated using 6.5 U1d and still same problem.

[daescobar]

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 :smileysad:

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

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

[James Sprinkle]

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]

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):

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

Regards,

[IBFsupport]

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

[LexCosta]

Hello,

I have exactly the same problem.

After upgrading the Windows 5.5 vCenter to VCenter 6.5 U2 Appliance most of the Hardware sensors stopped to work as it didbefore in vCenter 5.5.

We have diferente model of HPE Proliant Servers all with the same issue.

Did anyone get a solution for this problem already?

Thank you