Jakob Gottlieb Svendsen

Report
Kåre Rude Andersen
Operations Manager and the
Perfect world
The guy on stage
•
•
•
•
Chief System Management Architect & Co-founder of Coretech
SCOM since MOM 2000
Automation / Management Packs
Author of Mastering and Advanced SCOM Training.
•
•
•
•
Kåre Rude Andersen
Partner, Coretech A/S
[email protected], @kracoretech
blog.coretech.dk/kra
Agenda
•
•
•
•
•
•
•
•
•
SCOM – Love/Hate relationship
SCOM – Why it sucks
SCOM – Why its great
Automation – Alerts
Automation – Notification
Automation – Maintenance mode
Automation – SMA
Automation – Groups
Monitoring and Automation - Everything
SCOM – Love/Hate relationship
•
•
•
•
•
•
•
•
Lets do it – its a part of our License
Next next , dada
Import All Management Packs from Microsoft – di daaa di daaa
We did it – now what
YES We can see it all
Damm’it we can see it all
Alerts and Critical Health States
Bad reputation – often because of no processes in the company
Why it sucks
•
•
•
•
•
•
Deliver proactive information
Discovers everything you didnt knew you had
No control on Alerts
What, how and to whom do we need to present status
Do we really need to build applications topologies
Currently Microsoft are sleeping on features
SCOM – Why its great
•
•
ITEL / MOF / Six Sigma, VAL-IT, ISO 20000, SAS-70, etc
3 type of MPs:
–
–
–
•
•
•
Microsoft
SW/HW Vendor
Yourself
Totally open interface – you are in control
Working together with the rest of the suite
2012 R2 is stable and (fast)
SCOM – Solution
•
The solution is pretty simple:
• Implement Processes and Automate
SCOM – Solution – But what processes?
SCOM – Solution – But what processes?
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Procedure and script for installation and configuration of gateway server (depending on number of agents)
Procedure and script for Installation and configuration of agents to communicate with gateway (depending on step 1)
Procedure and script for enabling failover for agent communicating with more than one server.
Procedure and script for Installation and configuration of agents to communicate with a Management Server
Process to create groups for customers servers, either by registry or NetBIOS name
Process for Network monitoring
Process for Linux monitoring
Process and Script for creating Management Pack and naming conventions
Procedure for creation of Distributed Applications/Groups for Services
Procedure for Defining SLA/SLO on customers Services
Procedures for Rules/Monitor
Procedure for Presentation of reports – performance, availability etc.
Procedure for creation of customer views
Procedure for Subscriptions
Procedure for creation and configuration of Run As Accounts
Procedure for overrides
Procedures for Customer access to Reports, Views and Dashboards
Procedure for GSM Monitoring for Customers
Procedure for OpsInsights implementation
Configure Coretech SCOM Dashboard
Implement “Application groups” of servers – like Application, Web and Database.
Implement a gold, silver, bronze support environment by the use of registry.
Create Ticketing with your Helpdesk environment
Automation - Alerts
•
•
•
Please automate Alerts
Alerts could be self manageble
Always do a decision:
–
–
–
•
Disable it
Override it
Fix it
Introduce a 10 mins weekly Alerts meeting
Automation – Alerts
Manage Resolution State
$Alerts = get-scomalert -ResolutionState 0
if ($Alerts)
{
foreach($Alert in $Alerts)
{
$newState = $null
switch -wildcard ($Alert.Description)
{
"*Active directory*" { $newState = 10 }
"*DNS*"
{ $newState = 30 }
"*Cisco*"
{ $newState = 250}
}
switch -wildcard ($Alert.Name)
{
"ACME.Monitor.Event.Dummy.100" { $newState = 249 }
}
if($newState -ne $null)
{
$Alert.ResolutionState = $newState
$Alert.Update(“Resolution State changed automatically by the QueueManager Robot”)
}
}
}
Automation – Alerts
Daily Alerts
#Define last 24 hours
$AlertDateYesterdayBegin = [DateTime]::Today.AddDays(-1)
$AlertDateYesterdayEnd = [DateTime]::Today.AddDays(-1).AddSeconds(86399)
#Get alerts from last 24 hours
$YesterdayAlerts = @(get-scomalert | where {$_.TimeRaised -gt $AlertDateYesterdayBegin -and $_.TimeRaised -lt $AlertDateYesterdayEnd})
#Output
write-host
write-host
write-host
write-host
write-host
write-host
write-host
write-host
NUMBER OF ACTIVE ALERTS YESTERDAY: ($YesterdayAlerts).Count
CURRENT
CURRENT
CURRENT
CURRENT
NUMBER
NUMBER
NUMBER
NUMBER
OF
OF
OF
OF
ACTIVE
ACTIVE
ACTIVE
ACTIVE
ALL
CRITICAL
WARNING
INFORMATIONAL
ALERTS:
ALERTS:
ALERTS:
ALERTS:
@(get-scomalert
@(get-scomalert
@(get-scomalert
@(get-scomalert
|
|
|
|
where
where
where
where
{$_.ResolutionState
{$_.ResolutionState
{$_.ResolutionState
{$_.ResolutionState
-ne
-ne
-ne
-ne
‘255’}).count
‘255’ -and $_.Severity -eq ‘2’}).count
‘255’ -and $_.Severity -eq ‘1’}).count
‘255’ -and $_.Severity -eq ‘0’}).count
-foregroundcolor “red”
-foregroundcolor “yellow”
TOPLIST OF YESTERDAYS ALERTS SORTED BY COUNT:
#list and sort yesterday alerts
$YesterdayAlerts | Group-Object Name |Sort -desc Count | select-Object Count, Name |Format-Table –auto
write-host
#list and sort current active alerts
write-host CURRENT ACTIVE CRITICAL ALERT LIST: -foregroundcolor “red”
(get-scomalert | where {$_.ResolutionState -ne ‘255’ -and $_.Severity -eq ‘2’} | Group-Object Name |Sort -desc Count | select-Object Count, Name |Format-Table –auto)
Automation - Alerts
•
Introduce Resolution states like:
–
–
–
–
–
•
Handle by Service Owner
Awaiting Weekly Meeting
Do disable this Alert
Do override this Alert
Investigate this Alert
Reuse solution history with:
–
–
–
–
SCSM
ServiceNOW
Remedy
etc
Automation - Notification
•
•
•
•
•
Do not send Alerts directly to people
Always use Distributed Application or Groups as a source
Classify your Applications – Gold, Silver and Bronze
Prioritize your Applications – High, Medium, Low
Only send Alerts => Silver and High
Automation – Notification Alerts
Resend Alerts
$oldAlerts = Get-SCOMAlert | Where-Object {($_.LastModified -ge
[DateTime]::Now.AddHours(-4)) -and ($_.ResolutionState -eq 0)}
ForEach($alert in $oldAlerts)
{
$alert.Update("")
}
Automation – Alerts
Microsoft Alert Update Connector
DEMO - Automation – Alerts
Get more info into an Alert
$machine = Get-SCOMAlert | where {$_.ResolutionState -eq '110' -and $_.MonitoringObjectDisplayName -ne $()} |
ForEach { $_.ID
$AlertID = $_.ID
$strComputer = $_.NetbiosComputerName
$strFilter = "(&(objectCategory=computer)(objectClass=computer)(cn=$($strComputer)))"
$objComputer = ([adsisearcher]$strFilter).FindOne()
$Desc = $objComputer.properties.description
$alert = Get-SCOMAlert | where {$_.Id -eq $AlertID}
$alert.CustomField1 = ("Info from AD: " + $Desc)
$alert.ResolutionState = 0
$alert.Update("Got description from AD")
}
Automation – Notification
Catches
•
•
•
Do not send empty values = nothing happens
Remember the Notification Account
Notification Ressource Pool
Automation Maintenance mode
•
Tim Mcfadden Maintenancemode script
–
•
Stefan Stranger
–
•
•
https://gallery.technet.microsoft.com/scriptcenter/Put-OM2012-Computer-Group-43902672
Remember the catch about getting out of MM
–
•
http://www.scom2k7.com/scom-2012-maintenance-mode-scheduler/
Get-SCOMClass -name "Microsoft.SQLServer.2012.database" | GET-SCOMClassInstance
Orchestrator / SMA / SCCM
Christopher Keyaert SMA – Start SCOM MaintenanceMode
–
https://gallery.technet.microsoft.com/scriptcenter/SMA-Runbook-Start-SCOM-c594b92c
Automation with SCOM
Name
Runs
Benefits
Limitations
Orchestrator
On a Local server
Graphical Interface, Runs
all .Net Scripts
Max number of runbooks
running.
Service Management
Automation
1 or more Windows Azure Access any resource in
Pack server
your Datacenter –
Manage public/private
cloud, do pay $ for I/O
Interface (not really a
limitation)
Azure Automation
Azure public cloud
Cannot access your local
SMA or Orchestrator
(All) ways up
Automation - SMA
•
%windir%\system32\windowspowershell\v1.0\powershell.exe
•
-command "& {Start-SmaRunbook -WebServiceEndpoint "https://scwap" Port 9090 -Name "SCOM-RecoverHelloWorld" -Parameters
@{'Message'='Hello World!'}}"
Demo
•
Temperature – Warmer and Colder
SCOM – Using Recovery Action
Automation – Distributed Applications
•
•
•
Factfinder
Create automatically DA’s
Use as detective tool
Automation – Distributed Applications
•
Demo FactFinder
Automation - Everything
•
Logical Objects
–
–
Document Workflows, see status of batch
Show Health from a logical object
Automation - Everything
•
Physical Objects
–
If you can reach it you can control it
Automation - Everything
•
•
•
Logical Objects
Demo – Control Watt and Temperature
Create Logical Objects for Internet of things
THANK YOU
Send me an email for information on
Dashboards
Logical Objects
SCOM Health Check
Management Packs etc.
[email protected]

similar documents