Since SAM 6.3 supports installing SolarWinds Orion agents on Linux-based computers, you can use automation technologies like Chef or Puppet to mass deploy them.  The examples show two different scripting methods to deploy agents. These methods are independent of Chef or Puppet and either method can be used to automate your agent installations if it is supported by your automation software. And a special thank you to Toby.

By the way, SolarWinds doesn't support these automation technologies, and the instructions are provided as reference only and assume that your automation software is installed and configured correctly.

Here's what we've got:

Deploy an agent with an expect script

This method uses a Chef cookbook to create and deploy an expect script to target nodes. It contains two parts that must be performed in order.

Here's the example recipe used in the instructions:

# Install the swiagent package using the repository
    package 'swiagent'
# Create a driver to configure the agent
    cookbook_file '/opt/SolarWinds/Agent/bin/driver.ini' do
    source 'driver.ini'
    owner 'swiagent'
    group 'swiagent'
    mode '0755'
    action :create
       end  
# Start the agent and configure it 
# You only have to start and configure the agent once. 
# Assume that if the swiagent.cfg file exists, the agent is provisioned
    execute 'swiagent init' do
    command 'cat /opt/SolarWinds/Agent/bin/driver.ini | service swiagentd init'
    creates '/opt/SolarWinds/Agent/bin/swiagent.cfg'
  end

1. Create a recipe.
   1. Configure the node to use the SolarWinds repository.
   2. Get the package from the repository.
      

       package 'swiagent'

   3. Create the driver.
      

      cookbook_file '/opt/SolarWinds/Agent/bin/driver.ini' do
          source 'driver.ini'
          owner 'swiagent'
          group 'swiagent'
          mode '0755'
          action :create
       end

   4. Start and configure the agent.
      

      execute 'swiagent init' do
          command 'cat /opt/SolarWinds/Agent/bin/driver.ini | service swiagentd init'
          creates '/opt/SolarWinds/Agent/bin/swiagent.cfg' 
      end

   5. Save the recipe.
2. Create a driver.ini file that configures the agent to connect to the poller and use a specific Orion credential.
   

   2
   IPaddress   
   4  
   admin  
   7

You can apply the cookbook to any Chef registered node that you want to monitor with an Orion agent.

Deploy an agent with a shell script

This method uses a Puppet manifest to deploy a shell script to target nodes. It contains three parts that must be performed in order on the Puppet master node.

1. Create a module with an swiagent class.
   

   class swiagent {
       file { 'swi-driver':
       path => '/tmp/driver.sh',
       ensure  => 'present',
       replace => 'no', # always include this resource declaration
       mode    => '0644',
       source => 'puppet:///modules/swiagent/driver.sh',
       } ->
       exec { 'provision_agent':
       command => 'sh /tmp/driver.sh',
       path    => '/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin',
       creates => '/opt/SolarWinds/Agent/bin/swiagent.cfg', 
       } 
   }

2. Create a driver.sh file located in swiagent/files/.
   1. Click All Settings > Agent Settings > Download Agent Software.
   2. Click Linux, and Next.
   3. Click Manually Install by Downloading Files via URL, and click Next.
   4. Select your Distribution and Communication Mode
   5. Enter the Connection Settings.
   6. Click Generate Command.
   7. Copy this command to the driver.sh file.
      

      bash -c 'x(){ P=/Orion/AgentManagement/DownloadLinuxOnlineInstallScript.ashx? 
      requestId=eeaeac49-75c7-46bb-83b8-ee22c4924ef1;U=(http://host.local:8787/
      http://host:8787http://[2001:10:110:6:c802:e60b:70ba:7495]:8787  
      http://192.168.10.143:8787);dt(){ D=(wget curl);A=(-O\ -\ --no-check-certificate\  
      --tries=1\ --read-timeout=30 --insecure\ --retry\ 1); for((i=0;i<${#D[@]};i++));do  
      which ${D[$i]}&>/dev/null&&export DT="${D[$i]} ${A[$i]}"&&return;done;>&2 echo  
      "Cannot find download tool - please install some (${D[*]}) or use other  
      installation method";exit 1;};dt;echo export DT=\"$DT\";>&2 echo; >&2  
      echo -n "Downloading installation data from Orion Poller...";for u in ${U[*]};do  
      echo "export URL=\"$u\""; if ${DT} $u$P 2>/dev/null; then FOUND=1; break; fi; >&2  
      echo -n "."; done; >&2 echo ""; if [ "${FOUND}" != "1" ]; then >&2 echo "Unable to  
      connect to Orion Poller to download agent package. Please use an alternate deployment  
      method, such as Add Node wizard."; >&2 echo "See http://www.solarwinds.com/documentation
      /helpLoader.aspx?lang=en&topic=OrionAgentDeployAgentTop"; exit 1; fi; };X=$(x);bash -c "${X}"'

3. Add to the following to the main manifest for each node you want to monitor with an Orion agent:
   

   node 'host.local' { 
     include swiagent
     }

The next time the node pulls its configuration from the master, it evaluates the main manifest and apply the swiagent module.

Agent configuration options

These options may be different than the latest options built into the agent. Log on to the computer with the agent installed on it and type service swiagentd init to ensure that these options match the latest version of the software.

0. Cancel and Exit without Saving
1. Agent Mode (1 - Agent Initiated, 0 - Server Initiated)  [1]
2. Orion Poller Hostname/IP  [10.110.6.154]
3. Orion Poller Port        [17778]
4. Orion Username            [admin]
5. Orion Password            []
6. Proxy Settings...        [mode=disabled]
7. Save Changes and Exit

Let the community know how impressed you are with SAM and earn a $25 Amazon gift card!  Simply post your review in Spread The Word with the questions answered below:

1.       What was life like before using SolarWinds? (Include what you were using and why you decided it was time for a change)
2.        Which SolarWinds product(s) saved your bacon?
3.        Did you consider other options, and why did you choose SolarWinds?
4.       How has life been since you've rolled out SolarWinds in your environment?

I am planning to go through and remove some old custom properties that are no longer in use.  Is there an easy way (SQL Query, etc.) that can be done to look through all of my alerts for the use of any given custom property in the trigger conditions?  This would save me from having to go manually look at every alert I have configured.

Thanks in advance for any suggestions!

Hello!

I am trying to deploy an agent to a server across our WAN.

Ports that I have opened are 135 and 443 to the server and 17778 back as per the documentation.

This keeps failing with "Credentials test failed. Path not found" 

Analysing the network we find that it is trying to connect with ports, 445, 139 and 137

Are ports ports, 445, 139 and 137 the ones I should be opening?

Are there any other ports I should be using?

I having a tough time getting my script to correctly report the exit code to SAM 6.1.1. I have read The Basics of PowerShell (part 3) and I am using the required Message, Statistic and Exit codes. My script Telnets to a server (using a built telnet function, grabs a payload and verifies that its valid. I want 3 exit codes, 0 - Payload is confirmed (Up), 3 - Payload is invalid (Critical), 1 - Any error & failed to connect (Down). Exit codes 0 and 3 report correctly but 1 will not.

$error.clear()

$ErrorActionPreference= 'silentlycontinue'

#This builds a custom telnet function from http://community.spiceworks.com/scripts/show/1887-get-telnet-telnet-to-a-device-and-issue-commands

Function Get-Telnet

{  Param (

        [Parameter(ValueFromPipeline=$true)]

        [String[]]$Commands = @(""),

        [string]$RemoteHost = "",

        [string]$Port = "",

        [int]$WaitTime = 1000,

        [string]$OutputPath = ""

    )

    #Attach to the remote device, setup streaming requirements

    $Socket = New-Object System.Net.Sockets.TcpClient($RemoteHost, $Port)

    If ($Socket)

    {  $Stream = $Socket.GetStream()

        $Writer = New-Object System.IO.StreamWriter($Stream)

        $Buffer = New-Object System.Byte[] 1024

        $Encoding = New-Object System.Text.AsciiEncoding

        #Now start issuing the commands

        ForEach ($Command in $Commands)

        {  $Writer.WriteLine($Command)

            $Writer.Flush()

            Start-Sleep -Milliseconds $WaitTime

        }

        #All commands issued, but since the last command is usually going to be

        #the longest let's wait a little longer for it to finish

        Start-Sleep -Milliseconds ($WaitTime * 4)

        $Result = ""

        #Save all the results

        While($Stream.DataAvailable)

        {  $Read = $Stream.Read($Buffer, 0, 1024)

            $Result += ($Encoding.GetString($Buffer, 0, $Read))

        }

    }

    Else 

    {  $Result = "Unable to connect to host: $($RemoteHost):$Port"

    }

    #Done, now save the results to a file

    $Result | Out-File $OutputPath

}

#This clears the content of the output file so its not mistakenly read from previous telnet test

Clear-Content "F:\scripts\EDITelnetOutput.txt"

#This will telnet to the remote server and issue the command to check the payload. The output must be written to a file because get-telnet cmdlet sucks.

Get-Telnet -RemoteHost ${IP} -Port "3575" -Commands "Blah" -OutputPath "F:\scripts\TelnetOutput.txt"

$TelnetOutput = get-content "F:\scripts\TelnetOutput.txt"

#This looks for "Keyword" in the payload. If found then status is 1 and application is considered functional.

IF ($error)

{

Write-Host "Statistic: 1"

Write-Host "Message: $($error[0])"

    Exit 1

  }

IF ($TelnetOutput -match "Keyword") {

        Write-Host "Statistic: 0"

        Write-Host "Message: Valid Payload"

      Exit 0

    }

    ELSE {

        Write-Host "Statistic: 3"

        Write-Host "Message: Invalid Payload"

        Exit 3

          }

Tests for Exit 0 and 3 are successful with proper output. Test for exit 1 is done by turning off the application. The port is closed and the telnet connection will fail. Below is the error that I get for Exit 1 "Get Output Failed". It should result in Exit 1 - Component is Down.

I found that if I changed "Exit 1" any other number value it will run successfully. Something is wrong with Exit 1.

I am new to the Solarwinds world but have several alerts setup in the Advanced Alert Manager in Orion. I am now trying to monitor an old fax server we have. I want to monitor the event logs for an event id and alert if less than two matches per hour are sent. However I do not know what option to select under "Type of Property to Monitor" under the Edit Alert section.

See attached.

Hope I am making sense and asking in the correct section. The manuals are rather large.

I've setup and event log monitor to watch for event id 4740 and that works properly.  I'm trying to figure out how to get the username of the locked out account in the message of the email alert.  ideas?

What is the difference?  I have a Windows 2008 R2 x64 virtual machine (VMware).  Orion NPM says that the CPU has not gone over 25% in the past day.  Orion APM, Windows 2003-2008 Template, says that % Processor Time has been riding about 75-80 % constant in the past day.  There are no alarms in NPM, but APM is in a near constant alarm state.

The virtual server is a vSphere 5, vCenter 5 server - 4GB RAM and 2 x CPU.

Any help is appreciated.

- Dave Claussen

I'm trying to use the  Windows PowerShell Monitor component (actually as part of the "SolarWinds Web Performance Monitor (WPM) Player" template) in the Remote Host Execution Mode.  The component attempts to connect using an IP instead of an FQDN, so this error is generated:

PowerShell script error. Connecting to remote server 172.10.10.31 failed with the following error message : The WinRM client cannot process the request. Default authentication may be used with an IP address under the following conditions: the transport is HTTPS or the destination is in the TrustedHosts list, and explicit credentials are provided. Use winrm.cmd to configure TrustedHosts. Note that computers in the TrustedHosts list might not be authenticated. For more information on how to set TrustedHosts run the following command: winrm help config. For more information, see the about_Remote_Troubleshooting Help topic.

If I specify HTTPS instead, then this error is returned:

PowerShell script error. Connecting to remote server 172.10.10.31 failed with the following error message : The server certificate on the destination computer (172.10.10.31:5986) has the following errors: The SSL certificate contains a common name (CN) that does not match the hostname.

Is there some tricky way to tell the component monitor to access via some Fully Qualified Domain Name instead, or is this a feature request that needs to be made?  I realize that could go about modifying the TrustedHosts setting on all of my different pollers, but we'd prefer the ability to not to and to be able to use HTTPS as it's available for all of our connections.

New to this community so hello all!

I'm running SAM 6.2.4, and I'm fairly adept at Powershell, but I have a small problem regarding script arguments when trying to convert one of my standalone scripts into a monitor script.

For testing, I've tried various syntax but none have been successful in trying to pull an argument from outside into the script. Below is the last attempt (which doesn't work).

I've tried various in the script body...

The script runs fine but I can't seem to find the correct syntax to parse the argument.

In the last example, the Output Result is simply 'Message:'

Also, what is the separator for multiple arguments?

Can someone ease my pain.

Thanks

I am very much extremely green when it comes to both Raspberry Pis and Linux (till now I've lived within my SQL/PowerShell/C# world).

I have an Adafruit DHT temp sensor attached to a Pi 3 B, Python 2.7.9.

When I place the following code into a .py file within /tmp and call it while in a putty session, I am returned a line containing the data I am trying to pull, but when I put the same script in a Linux SAM component, I get:

"Testing on target node failed with 'Down' status ('Down' might be different if script exits with a different exit code).

#!/usr/bin/python

import sys

import Adafruit_DHT

humidity, temperature = Adafruit_DHT.read_retry(11, 4)

print 'Temp: {0:0.1f} C  Humidity: {1:0.1f} %'.format(temperature, humidity)

notes:

Script Argument is just: python ${SCRIPT}

Script Working Directory: /tmp

I am using the pi account, credentials added in SolarWinds, of course.

additionally, I've discovered that it is bombing on the "humidity, temp....." line, not on the "import Adafruit_DHT" line

Why would the exact same code execute perfectly from the /tmp dir, logged in as "pi", yet fail when inside of Linux monitor component using same dir and credentials?

-Lost

I am new to the Solarwinds world but have several alerts setup in the Advanced Alert Manager in Orion. I am now trying to monitor an old fax server we have. I want to monitor the event logs for an event id and alert if less than two matches per hour are sent. However I do not know what option to select under "Type of Property to Monitor" under the Edit Alert section.

See attached.

Hope I am making sense and asking in the correct section. The manuals are rather large.

I having a tough time getting my script to correctly report the exit code to SAM 6.1.1. I have read The Basics of PowerShell (part 3) and I am using the required Message, Statistic and Exit codes. My script Telnets to a server (using a built telnet function, grabs a payload and verifies that its valid. I want 3 exit codes, 0 - Payload is confirmed (Up), 3 - Payload is invalid (Critical), 1 - Any error & failed to connect (Down). Exit codes 0 and 3 report correctly but 1 will not.

$error.clear()

$ErrorActionPreference= 'silentlycontinue'

#This builds a custom telnet function from http://community.spiceworks.com/scripts/show/1887-get-telnet-telnet-to-a-device-and-issue-commands

Function Get-Telnet

{  Param (

        [Parameter(ValueFromPipeline=$true)]

        [String[]]$Commands = @(""),

        [string]$RemoteHost = "",

        [string]$Port = "",

        [int]$WaitTime = 1000,

        [string]$OutputPath = ""

    )

    #Attach to the remote device, setup streaming requirements

    $Socket = New-Object System.Net.Sockets.TcpClient($RemoteHost, $Port)

    If ($Socket)

    {  $Stream = $Socket.GetStream()

        $Writer = New-Object System.IO.StreamWriter($Stream)

        $Buffer = New-Object System.Byte[] 1024

        $Encoding = New-Object System.Text.AsciiEncoding

        #Now start issuing the commands

        ForEach ($Command in $Commands)

        {  $Writer.WriteLine($Command)

            $Writer.Flush()

            Start-Sleep -Milliseconds $WaitTime

        }

        #All commands issued, but since the last command is usually going to be

        #the longest let's wait a little longer for it to finish

        Start-Sleep -Milliseconds ($WaitTime * 4)

        $Result = ""

        #Save all the results

        While($Stream.DataAvailable)

        {  $Read = $Stream.Read($Buffer, 0, 1024)

            $Result += ($Encoding.GetString($Buffer, 0, $Read))

        }

    }

    Else 

    {  $Result = "Unable to connect to host: $($RemoteHost):$Port"

    }

    #Done, now save the results to a file

    $Result | Out-File $OutputPath

}

#This clears the content of the output file so its not mistakenly read from previous telnet test

Clear-Content "F:\scripts\EDITelnetOutput.txt"

#This will telnet to the remote server and issue the command to check the payload. The output must be written to a file because get-telnet cmdlet sucks.

Get-Telnet -RemoteHost ${IP} -Port "3575" -Commands "Blah" -OutputPath "F:\scripts\TelnetOutput.txt"

$TelnetOutput = get-content "F:\scripts\TelnetOutput.txt"

#This looks for "Keyword" in the payload. If found then status is 1 and application is considered functional.

IF ($error)

{

Write-Host "Statistic: 1"

Write-Host "Message: $($error[0])"

    Exit 1

  }

IF ($TelnetOutput -match "Keyword") {

        Write-Host "Statistic: 0"

        Write-Host "Message: Valid Payload"

      Exit 0

    }

    ELSE {

        Write-Host "Statistic: 3"

        Write-Host "Message: Invalid Payload"

        Exit 3

          }

Tests for Exit 0 and 3 are successful with proper output. Test for exit 1 is done by turning off the application. The port is closed and the telnet connection will fail. Below is the error that I get for Exit 1 "Get Output Failed". It should result in Exit 1 - Component is Down.

I found that if I changed "Exit 1" any other number value it will run successfully. Something is wrong with Exit 1.

Hello all,

Does any one know what is the problem? Let's say I have a server monitored by SolarWinds APM, for example file server. And I want to be aware that I reached space issue for some volumes of the server.

I'm looking a node status and node has green light - means node ok.

Ok, let's scroll down a little bit to review node's volumes

all 9 volumes looks good, all is green. From that step I should close the page and be sure that i have enough free space on each volume, isn't it?

But when I move cursor on the specific volume  I can see that for example volume H has 89% used space and looks like in the warning state

Can some one explain to me the following:

     1. Why all the volumes are green even it has used space over warning threshold ?

     2. Why node state is normal once the node's component reached warning threshold ?

Hopefully everyone knows that the SAM 6.5 beta is available. If not, here is your official notice. Now is the time to head over to Server & Applications Monitor Beta and give us feedback on the new features. Feedback gets you THWACK points. THWACK points get you closer to the top of the leaderboard, and cool swag from the THWACK Store. What are you waiting for?

Just patched our SAM server which runs on Server 2008R2. After patching I saw a number of events related to a Solarwinds service and .net crashing. The service restarts just fine but crashes again in 5-10 minutes. Doesn't seem to impact anything except I'll occasionally get a page not found error if the crash has just occurred, once the service restarts everything is fine for a bit.

Event log info:

Task Category: None

Description:

Application: SolarWinds.BusinessLayerHost.exe

Framework Version: v4.0.30319

Description: The process was terminated due to an internal error in the .NET Runtime at IP 71BAC37A (71B00000) with exit code 80131506.

Event Xml:

<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

  <System>

  </System>

  <EventData>

Framework Version: v4.0.30319

Description: The process was terminated due to an internal error in the .NET Runtime at IP 71BAC37A (71B00000) with exit code 80131506.

</Data>

  </EventData>

</Event>

Task Category: (100)

Description:

Faulting application name: SolarWinds.BusinessLayerHost.exe, version: 2015.1.5500.2341, time stamp: 0x54e33230

Faulting module name: clr.dll, version: 4.0.30319.34209, time stamp: 0x5348961e

Exception code: 0xc0000005

Fault offset: 0x000ac37a

Faulting process id: 0x1590

Faulting application start time: 0x01d1814f54e5b28b

Faulting application path: C:\Program Files (x86)\SolarWinds\Orion\SolarWinds.BusinessLayerHost.exe

Faulting module path: C:\Windows\Microsoft.NET\Framework\v4.0.30319\clr.dll

Report Id: 3a3c487d-ed43-11e5-8e3d-ac162d8d6817

Event Xml:

<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

  <System>

  </System>

  <EventData>

  </EventData>

</Event>

We have set up the nodes for our VCSA (vCenter Server Appliance) and PSC (vCenter Platform Services Controller) in SolarWinds for monitoring.  However, the average CPU and memory statistics for both show 100%.  If you look at the resource consumption in vSphere neither the CPU or memory statistics are anywhere close to that.  According to vSphere performance monitoring our VCSA CPU average is 19.3% and the memory is 11% and the PSC CPU average is 2.8% and the memory is 13.7%.

We have rebooted both guests, removed and added back the resource from the node, and unmanaged/remanaged each node.  In SolarWinds the monitors for both guests and their resources are showing 100%.

Has anyone experienced this before?  If so, what did you do to get it fixed?

Current environment

VCSA and PSC are VMware Linux OS

ESX hosts are at 6.0.0, build 4600944

Orion Platform 2016.1.5300, SAM 6.2.4

For much longer than I care to admit I was struggling to monitor any Windows services using WMI on one of my Windows 2008 Servers. Instead I received the dreaded "Invalid Class" error in the APM info window. This was later confirmed using WBEMTEST. When using WMI Explorer it was clear to see that the "Win32_PerfRawData_PerfProc_Process" class did not exist. 

I opened a case with Solarwinds support and was directed to their WMI support document and related online documentation. Unfortunately none of the information provided proved useful in this circumstance.

I spent a several hours googling around and was able to find several others with my issue, but no one had found a solution. With nowhere else to go, I spent the $250.00 and called Microsoft Support. It took a few days, and at least three Microsoft support engineers later that we finally got to the root of the problem. I knew from the very beginning that this was going to be something stupid, but I never realized just how obscure the problem would be, so I decided to save others my pain and post my resolution in hopes that others might find it useful in the future. 

We went into registry “HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\PerfProc\Performance” In the right hand pane we found that “Disable Performance Counters” was set to 1, which means it was disabled. We used “exctrlst.exe” tool, enabled ‘perfproc’ and rebooted. After Rebooting we were able to run the WMI query (SELECT * from Win32_PerfRawData_PerfProc_Process where NAME <> '_Total') using the WBEMTEST tool built into Windows. After verifying with WBEMTEST I confirmed that APM was properly polling my monitored services properly. 

I hope someone else finds this helpful.

New to this community so hello all!

I'm running SAM 6.2.4, and I'm fairly adept at Powershell, but I have a small problem regarding script arguments when trying to convert one of my standalone scripts into a monitor script.

For testing, I've tried various syntax but none have been successful in trying to pull an argument from outside into the script. Below is the last attempt (which doesn't work).

I've tried various in the script body...

The script runs fine but I can't seem to find the correct syntax to parse the argument.

In the last example, the Output Result is simply 'Message:'

Also, what is the separator for multiple arguments?

Can someone ease my pain.

Thanks

So this just started today.....

Performing maintenance on nodes and we mute alerts, but the process failed (or so it says) but it will show as muted. When we try to unmute, the process failed and we cannot seem to get things back to normal. Anyone else having this issue?