SCOM 2019 – UR4

SCOM 2019 UR4 is here, more information can be found on https://support.microsoft.com/en-us/topic/update-rollup-4-for-system-center-operations-manager-2019-07ad0ef3-a330-4373-92f6-2dda3821bee or read below:

Download Update Rollup 4 for Microsoft System Center 2019

Install the update in the following order:

Management Servers

Web Console Servers

Gateway Servers

Operations Console Servers

Reporting Server

Apply Agent Updates

Update UNIX/Linux management packs and agents

Updated UNIX/Linux management packs can be found here

Improvements

  • Support for Windows 11
  • Enabled .NET4.8 support
  • UI improvements in Operations console:
    • Support for sort option by column, in Overrides Summary.
    • For Monitors, Rules, Task and Discoveries, Management Pack label text is selectable in the workflow Properties window.
    • Added new fields for Class Technical Name in the State Views. Added the same in the wizard for creating a new Alert, Event, Performance or State View.
    • Added Target Class Display Name to help identify the target of a rule while selecting rules during the creation of a new Performance View.
    • Added 3 new columns Management Pack, Sealed and Members in the Authoring pane > Groups.
    • Added new column for Management Pack Display Name in the Authoring pane > Create Group wizard.

Issues that are fixed

Operations Manager 2019 Update Rollup 4 includes fixes for the following issues:

Operations Console  

  • Get-SCOMGroup CmdLet in PowerShell not reflecting the updated name of a group .
  • Edited alert properties in the alert views not being reflected sometimes without switching views.
  • The Reason and/or Comment field not being updated correctly in an existing maintenance mode schedule.
  • Maintenance Mode Schedules fail to work when creating multiple schedules at once.
  • Information around update rollup applied to Management Servers not being reflected correctly in the Ops console Admin view.
  • MonitoringHost.exe process(es) crashes when a Management Server had neither Microsoft OLE DB Driver 18 for SQL Server nor SQL Server 2012 Native client 11.0 installed.
  • In high-load environments, when AlertUpdate data was put in the Pending send queue, there was an issue processing these and event 5204 would get logged which was causing the All-Management Resource Pool to go down.
  • Updating an existing Console Task would automatically switch the Task Target (class) to System entity instead of the one that was initially selected.
  • Edit an existing Active Directory Integration Failover Configuration when wanting to add or remove Management Servers in the Failover list.
  • The MP Reference Alias generator is case-sensitive now and will create only unique Aliases for Management packs.
  • The “Installed column” is now sortable in the Operations Manager product views.

Web Console

  • Authentication failing in web console if password of the account used would contain the “:” character.
  • New Dashboard Wizard in the web console would fail with MPInfra_p_ManagementPackInstall exception.

Other Fixes:

  • Database Registry that was created for temporary usage after patch or upgrade installation in 2019UR3 is now being deleted.
  • There was a special scenario where APM would crash IIS Application Pools when W3WP was trying to load any type of managed DLL which was built as 64bit. These types of DLLs will no longer cause APM to crash the IIS Application Pool.
  • The Insecure Direct Object Reference Vulnerability issue related to APM websites
  • SNMP Traps sent through a proxy would not be received.
  • Network discovery failure for network devices with HSRP that do not have the AF_NET IP address.
  • Generic event reports coming up empty for events generated by SNMP devices.
  • SDK service not being able to reconnect to SQL after a long SQL AG failover.

Performance improvements:

  • Optimized the table indexes to improve record insertion rate.
  • Added indices to 2 Data Warehouse tables to improve performance.
  • Included Date and time information for Performance Data collected by PowerShell scripts.
  • The PerformanceRuleInstance table which was causing performance issues is now being groomed correctly.
  • Improved performance of ACS report queries that are using the [dbo].[spInsertString] stored procedure .

Unix/Linux/Network monitoring fixes and changes:

  • The counter name “% Free Space” has been changed to “Used Megabytes” in Linux RHEL 6 and RHEL 7 MP for accurate representation.
  • Corrected the performance counter collected by “Used Megabytes” rules in RHEL 6 and 7 MPs.
  • The file system statistics have been enhanced to match the data with Linux standard commands.
  • Linux agent  now uses SHA2 to encrypt for certificates.
  • log rotate configuration files were getting deleted as part of Linux agent upgrade.
  • OMI agent crash issue during log rotate.
  • OMI logrotate SELinux module was getting uninstalled as part of Linux agent upgrade
  • Memory corruption issue that would cause a crash of MonitoringHost.exe when it would fail to perform a SSH call to a Cross-Platform Agent has been fixed.
  • ECDSA and ED25519 keys are now able to authenticate with Operations manager Console.
  • Some JBOSS applications were not discovered from JEE MP.
  • Fixed the /etc/opt/microsoft/scx/conf/scx-release file content for Oracle Linux
  • Increased the limit to 128 words (from 64 words) for the filter associated with OMI WQL queries.
  • Process monitoring related issues for Unix/Linux servers with a very large number of processes.
  • Privilege escalation security issue related to Kerberos setup.
  • Non-IP cluster resource (ex. Disk) getting discovered as an IPv6 resource.
  • TLS 1.2 is now enabled by default for Linux agents.
  • Cross platform DLLs hash mismatch is fixed.
  • Linux Agent memory leak issue.

Previous fixes

In addition to these issues, all the issues that are fixed in SCOM 2016 UR10 and earlier update rollups for SCOM 2019 are also fixed in SCOM 2019 UR4.

MSSQL on Windows: WMI Health State Errors

MSSQL on Windows: WMI Health State Errors

If you, like me – Have received a ton of these alerts – do not fear! There is a fix out there for you.

I will keep it short! There is an SQL Server MP update from Microsoft – Found here: https://www.microsoft.com/en-us/download/details.aspx?id=56203

What’s New

  • Updated MP to support SQL Server 2019 RTM
  • Added filter by edition to “Local DB Engine Discovery”
  • Redesigned DB Space monitoring to improve performance: Enabled by default monitors and performance rules targeting Database which watch for disk space consumption by ROWS Filegroups and Logfiles
  • Redesigned DB Space monitoring: Added two monitors and two performance rules targeting Database to watch for disk space consumption by In-Memory and FILESTREAM data
  • Redesigned DB Space monitoring: Read-only filegroups now count as well
  • Redesigned DB Space monitoring: Disabled by default all workflows targeting Filegroups, Files, Logfiles
  • Redesigned XTP performance counters to make them completely version-agnostic
  • Added attribute “TCP Port” to “SQL DB Engine Class” and updated “DB Engine Discovery” to populate the new property
  • Added summary dashboard for SCOM 2019 Web Console (HTML5)
  • Added support for cluster nodes with disjoined namespaces
  • Added sampling to algorithm of monitor “WMI Health State” in order to eliminate false alerting on cluster SQL Server instances
  • Updated alert descriptions of monitors “Availability Database,” “Availability Replica,” and “Availability Group” (generating alerts still disabled by default)
  • Updated monitor “Product Version Compliance” with versions of most recent public updates to SQL Server
  • Disabled by default monitor “Buffer Cache Hit Ratio” and changed its threshold from 0% to 90%
  • Disabled by default monitor “Page Life Expectancy”
  • Removed monitors “Availability Database Join State” and “Availability Replica Join State” as not useful
  • Updated display strings
  • Revised columns on DB Engine state views

Issues Fixed

  • Fixed: monitor “Service Principal Name Configuration Status” raises false alerts because of case-sensitive comparison
  • Fixed: “Local DB Engine Discovery” crashes when Windows has Turkish locale
  • Fixed issue that caused performance degradation in workflows “General Always On Discovery,” “Database Replica Discovery,” and “Always On System Policy Monitoring”
  • Fixed: “General Always On Discovery” throws errors on environments with several Distributed Availability Groups
  • Fixed monitoring issue in case of Database is replicated by Always On Availability Group
  • Fixed empty property bag when Availability Group has cluster type NONE
  • Fixed wrong target in alerting rule “DB Backup Failed to Complete”
  • Fixed rule “MSSQL Integration Services on Windows: The package restarted from checkpoint file” and its alert
  • Fixed rule “OS Error occurred while performing I/O on pages“ and its alert
  • Fixed: “DB Disk Write Latency” and “DB Disk Read Latency” monitors and performance rules get wrong performance metric
  • Fixed alert description of monitor “WMI Health State”

Microsoft & ServiceNow in a strategic partnership – What could it mean?

Microsoft & ServiceNow in a strategic partnership – What could it mean?

I’ve changed the header a few times, from what could it mean, to what will it mean and back to could… The press release points alot to Azure, but could this be a revival to System Center with more companies moving to a hybrid setup? Let me give you my view.

We all know that Microsofts ITSM tool ‘System Center Service Manager’ was not the tool that Microsoft wanted it to be. While having integrations to all the other System Center tools, Service Manager was a tool used of necessity more than the features the tool provided.

With this announced partnership, it is a conclusion from my end that Service Manager will be ‘put down/downgraded’, and integrations to all the other System Center tools will be made available officially from Microsoft & ServiceNow. Even though it does not take a whole lot to set up the connector between ServiceNow and OpsMgr, I do think that this will be an integrated part of the tools from the next annual patch.

It is my hope that more of the SCOM reporting will be made available through, and more of the collected data available out of the box. There is alot of reasons why this could be a good approach, one of them being that ServiceNow reporting is much easier to understand than the SCOM part since most people using the reports are not SCOM admins.

Workflows for user, server, workstation, service, application, deployment creation is another thing that could also work ‘out of the box’. Instead of using hours on creating workflows in ServiceNow, and then often in Powershell, this could be made available directly from the companies with full support instead of hiring third party consultants.

Tinfoil hat prediction…? Microsoft to acquire ServiceNow? I could see it happen.

Monitor ‘Health Service State’ folder for changes

Monitor ‘Health Service State’ folder for changes

Every now and then, SCOM fires away a powershell script that fails – and sometimes the error message you receive is not very useful. To make matters worse, the script will then delete it self so the chances of you to troubleshoot this – is limited.. Until now

This script will allow you to monitor the Health Service State folder (or any other folder, if you change the variable) for powershell scripts (or any other extension, if you change the variable). While this script is running, it will copy any of the desired extensions over to another folder (you set this up in the variable) for you to troubleshoot.

You can view/copy/download the script right here

Disclaimer: 
This script has been used in a production environment, but may differ from yours. RSG takes no responsibility for any damage this script may cause to your system.