none
SCOM 2016 - Agents 1.6.2-337 - Module was unable to convert parameter to a double value

    Question

  • Hi,

    I have SCOM 2016 with last Unix/Linux MP. All agents have last version.

    All management servers have alerts:

    Operations Manager Failed to convert performance data

    Module was unable to convert parameter to a double value

    Original parameter: '$Data/Value$'

    Parameter after $Data replacement: '1.#QNAN'

    Error: 0x80020005

    Details: Type mismatch.

    One or more workflows were affected by this.

    Workflow name: Microsoft.Linux.Universal.NetworkAdapter.BytesTotalSec.Collection

    Instance name: eth2

    Instance ID: {BC351E66-8BB4-31BB-E576-AF2B5FF11369}

    Thursday, January 26, 2017 11:19 AM

All replies

  • Hi,

    Can you please check, if you are on the latest UR1 installed on Operations Manager machine.

    I would like you to re-check, if you have correct MP (Universal Linux Monitoring) with version 7.6.xxxx.0 installed on the Operations Manager 2016.

    According to me you have the incorrect MP (Universal Linux Monitoring) installed, which is for SCOM 2012 R2 and has version 7.5.xxxx.0, which is not compatible with SCOM 2016.

    Thanks,

    Nirbhay Singh



    Sunday, January 29, 2017 7:09 PM
  • Hi,

    All MPs are 7.6.xxxx.0.

    Monday, February 06, 2017 10:51 AM
  • Hi,

    Can you please file a support ticket with Microsoft and please add the URL for this post in the same.

    Thanks,

    Nirbhay Singh

    Monday, February 06, 2017 3:35 PM
  • Can you please take traces for SCOM agent on Management Server :

    Steps to get traces for SCOM Agent : http://blog.scomskills.com/scom-verbose-tracing/

    Attach the same to the support ticket.

    Thanks,

    Nirbhay Singh

    Monday, February 06, 2017 4:54 PM
  • I'll wait for UR2. If update rollup won't fix this issue I'll try to open a support ticket.
    Friday, February 10, 2017 7:25 AM
  • Hi,

    We have recently updated our OM 12R2 to OM 16 and got the same issue.

    Some troubleshooting actions gave us the answer to the question "why?"

    First, I've checked a few event IDs and noticed that the warnings are always about some interfaces only. Lucky me, or just some linuxes are affected?

    Let's find that out:

    Import-Module operationsmanager
    $guid = @()
    $event = Get-WinEvent -FilterHashtable @{Logname='Operations Manager';Id=11052} -MaxEvents 3000
    $event | % {
     	$xml = [xml]$_.ToXml()
    	$guid += ((($xml.Event.EventData.Data)[3]).Replace('{','')).Replace('}','')
    }
    $guid | select -Unique | % {
    	get-scomclassinstance -id $_ | Select id, Path, DisplayName
    }


    OK, I can see only a few linux servers (5 actually), but each of them has two or more interfaces:

    Hmm, maybe thats the reason? Let's check that! Hopefully, I have the same management account for all the servers in that management group:

    $Cred = Get-Credential
    $linuxagents = Get-SCXAgent | sort name | select name
    $linuxagents.Name | % {
        Write-Host ($_ + ': ' + ((Get-WSManInstance -computer $_ -authentication basic -credential $cred -port 1270 -usessl -enumerate http://schemas.microsoft.com/wbem/wscim/1/cim-schema/2/SCX_EthernetPortStatistics?__cimnamespace=root/scx | measure).count))
    }

    Hey, all the servers with at least two interfaces are affected!


    Moreover, how about the performance rules? Here is the normal server:



    Here is the server with more than 1 interface:

    10^16 bytes/sec? 13 Petabytes per second, really?


    In addition, here is the real output for a normal server:

    Here is a real output for the server with a few interfaces:


    Looks like that if you have only the one interface, the measurements are correct (i.e. speed = current value - previous value). Nevertheless, when we have a few interfaces, the speed just equals the current value.







    • Edited by Alexx_B Monday, February 20, 2017 2:56 PM
    Monday, February 20, 2017 2:34 PM
  • Same issue here, even with agent version 1.6.2-338 from UR2.
    Friday, February 24, 2017 12:50 PM
  • Can you please update the SCOM with UR2. One of our customers have reported updating SCOM UR2 fixed the fix.

    Thanks,

    Nirbhay Singh

    Friday, February 24, 2017 5:15 PM
  • Nirbhay, I still have same issue with SCOM 2016 UR2 and Linux-agents 1.6.2-338.
    Tuesday, February 28, 2017 8:50 AM
  • We see the same issue in all installations running SCOM 2016 UR2 and Linux-agents 1.6.2-338. Must be a bug in the MPs.

    Event ID is 11904 (error) and event ID 11052 (warning) is logged for machines with more than one network interface

    11904:

    The Microsoft Operations Manager Expression Filter Module failed to query the delivered item, item was dropped.
    Property Expression: $Data/WsManData/*[local-name(.)='SCX_EthernetPortStatistics']/*[local-name(.)='BytesReceived']$
    Error: 0x80004005
    One or more workflows were affected by this. 
    Workflow name: Microsoft.Linux.RHEL.4.NetworkAdapter.BytesReceivedPerSec.Collection
    Instance name: bond0
    Instance ID: {FD22D458-8A22-BDF7-9CEC-EA2724DF97F7}

    11052:

    Module was unable to convert parameter to a double value
    Original parameter: '$Data/Value$'
    Parameter after $Data replacement: '1.#QNAN'
    Error: 0x80020005
    Details: Type mismatch.
    One or more workflows were affected by this. 
    Workflow name: Microsoft.Linux.RHEL.7.NetworkAdapter.BytesReceivedPerSec.Collection
    Instance name: eno16780032
    Instance ID: {22F8551C-7F69-4E21-6EEB-85A29BFA0DA0}

    Br

    Lars

    Wednesday, March 01, 2017 8:42 AM
  • Please open a support case with Microsoft so this issue can be properly debugged and documented.

    https://support.microsoft.com/en-us/premier

    https://support.microsoft.com/en-us/premier/contacts

    1-800-936-3100

    Tuesday, March 07, 2017 4:50 PM
  • Did anyone open a Support case yet? If so, please let me know the Service Request ID, we're thinking about joining then since we have the same issue.
    Wednesday, March 22, 2017 12:05 PM
  • I've opened a case and its already closed now. The Problem will be fixed in UR3 (expected in may 2017).

    #117032015476237

    Thursday, March 30, 2017 6:58 AM
  • Same problem here with Suse 11+12 and Redhat 7 MPs on Agent 1.6.2-338. Until UR3 we disable the specific collection rules:

    Microsoft.Linux.SLES.12.NetworkAdapter.BytesTotalSec.Collection

    Microsoft.Linux.SLES.12.NetworkAdapter.BytesSentPerSec.Collection

    Microsoft.Linux.SLES.12.NetworkAdapter.BytesReceivedPerSec.Collection

    Microsoft.Linux.SLES.11.NetworkAdapter.BytesTotalSec.Collection

    Microsoft.Linux.SLES.11.NetworkAdapter.BytesSentPerSec.Collection

    Microsoft.Linux.SLES.11.NetworkAdapter.BytesReceivedPerSec.Collection

    Microsoft.Linux.RHEL.7.NetworkAdapter.BytesSentPerSec.Collection

    Microsoft.Linux.RHEL.7.NetworkAdapter.BytesReceivedPerSec.Collection

    Microsoft.Linux.RHEL.7.NetworkAdapter.BytesTotalSec.Collection

    Thursday, April 06, 2017 2:49 PM
  • Joining the party. We experience the same behaviour.
    Wednesday, April 26, 2017 9:40 AM