First Look at OpsLogix’s VMware Management Pack (beta)

Yep you’ve heard right OpsLogix’s team is putting their forces on creating Management Pack for monitoring VMware. This blog post will not be comparison with another well known management pack for VMware. I will try to give you overview of the VMware Management Pack that OpsLogix is cooking up. Keep in mind that the MP is still in beta. Although the MP is in beta the OpsLogix’s support was not in beta. While I was testing the MP I’ve stumbled on a bug. I’ve contacted OpsLogix’s support, granted them access to my environment and after a couple of hours I’ve received a fix. Such quick fix shows the devotion of OpsLogix on supporting their products. Another reason to try this MP is that I haven’t touched vCenter since I switched to Hyper-V and System Center (somewhere around 3-4 years ago).

Before importing the MP let’s first see the environment where I am testing:

SCOM server named SCOM01. The SCOM version is 2012 R2 Preview. The operating System is Windows Server 2012 R2 Preview.
vCenter server named vcenter. Version is 5.1. Operating system is Windows Server 2012.
Two ESXi hosts 192.168.100.150 and 192.168.100.152 that are in cluster:

First you need to import two management packs:

After successful import you need to configure the resource pool that the MP will use. On importing OpsLogix VMware MP creates its own a resource pool:

You need to make the pool manual and add the SCOM management servers that will be used for monitoring your VMware environment.

In my case I have only one SCOM management server but in real world environment you will have more than one and you can distribute your resources more equally:

It is good that the MP is taking advantage of Resource Pools to distribute the load.

Before starting monitoring of the VMware environment let’s see with MPViewer what classes, discoveries, groups, relationships, monitors and rules are offered:

Classes

Discoveries

Groups

Monitors

Relationships:

Rules:

We will go in detail how the MP works but let’s first configure it to monitor the VMware environment.

You you will need to go to Monitoring pane -> OpsLogix –> VMWare –> VMWare IMP Configuration Dashbaord:

There you need to enter the name of your vCenter server and credentials. You do not need to have SCOM agent installed on that server to activate monitoring. The credentials provided can be domain account or local account on your vCenter server you just need to grant that account read-only permissions to the root of your vCenter server:

There is a button that will allow to test your connection before adding it:

If connection is successful you a re good to go:

So how the MP works?

The MP connects to the vCenter web service. By connecting to it it will find all objects (datacenters,ESX servers, clusters, datastores, vmnetworks and etc.) in your vmware environment:

For generating alerts the MP uses the built-in functionality in vCenter to create alarms. If you create an alarm in vCenter and that alarm is triggered it will show in SCOM also.

If you close an alert in SCOM that was not resolved in vCenter that alert will appear again in SCOM. Unfortunately if you close alarm in vCenter it will not be closed in SCOM. May be in some future version of the MP we will see option to enable synchronization from both side depending on your preference.

Besides the alerts the MP also tracks state of the different object. Depending on the alarms that are generated or not generated for particular object OpsLogix VMWare MP will calculate different states (healthy, warning or critical) on small intervals.

So the MP is very customizable and customizations can be made straight from vCenter which I think is the preferable way for vmware administrators and will save money from your budget if you had to teach them to do that in SCOM.

These are the capabilities of the MP from first sight but let’s see how it can provide more value by mixing it with more SCOM magic.

What about Diagram of your VMWare environment? Something what VMM MP has.

Sure that can be done you just need to create new diagram view:

And Viola:

What about VMWare Dashboard? I want to track the state of my VMWare environment in a dashboard.

A few simple steps and you are ready:

OK we’ve showed what are the capabilities of the MP now let’s show what the current beta version cannot offer otherwise it wouldn’t be a fair review. Performance alerts can be triggered trough configuring alarms in vCenter but the MP itself does not gather performance data to put it in SCOM databases. Probably because of this no report is available also. For such capabilities you have to look for solution at vCenter side.

Summary

OpsLogix VMWare MP looks promising at its first steps and it is definitely worth trying it if you consider monitoring your VMWare environment with SCOM. OpsLogix have proven that they are good at making MPs. The MP does not have price yet as it is in beta but it is advertised by OpsLogix as MP that make sense in price and performance. If you want to try it you can find more information here.

Testing OpsLogix Oracle Intelligent Management Pack

OpsLogix is a Microsoft System Center Alliance partner which develops innovative products for (virtual) infrastructure management. In my experience I was previously working with Oracle Intelligent Management Pack by supporting the SCOM environment of one of OpsLogix reference customers. I’ve never had issues with that MP. As being only supporting part I was not involved in implementing this MP but recently I was given this opportunity so this blog post is about my experience with OpsLogix Oracle Intelligent Management Pack.

First I want to start that instead of testing the Oracle IMP in supported scenario I’ve decided to test in unsupported environment in order to prove how good this MP is. I’ve always been enthusiast so I just I couldn’t resist. Also I want to make a notice that if you are deploying in production always use supported scenario.

Before introducing my test environment let’s first see what the architecture of the Oracle IMP is:

As we can see you just has to have one Windows node with SCOM Agent on it and ODAC.Net components installed. That node acts as a proxy and connects to all you Oracle database weather they lie on Windows OS, Unix OS or Linux OS. This architecture makes the MP very flexible which is especially important in the changes about unix an linux agents changes coming in SCOM 2012 R2. Basically it makes the MP SCOM 2012 R2 ready.

Now let’s look of how my lab looks:

· SCOM 2012 R2 Preview server that is installed on Windows Server 2012 R2 Preview.

· Oracle ODAC.Net components installed on the SCOM 2012 R2 Preview Management Server. You can install them on separate node but I’ve used my SCOM management server for simplicity. The components require .NET Framework 2.0 and are supported on Windows 2003/2008/2008R2 with x86 or x64 architecture. In production install them only on supported systems. I am sure that in the future OpsLogix will offer support for WS 2012 also.

· Windows Server 2012 (DB1) with Oracle 12c installed on it. Oracle 12c is not supported by the Oracle IMP but I’ve wanted to see if it is possible to add such database.

· Windows Server 2012 (DB2) with Oracle 11g installed on it.

After the overview of the lab we can continue with the first steps of implementing Oracle IMP.

1. First make sure you’ve installed .NET Framework 2.0 on the node that you will use as proxy. I already had it installed on my SCOM 2012 R2 Preview management server.

2. Second install SCOM agent on that same server. As my server is management server I do not need to do that.

3. Next you can proceed with installing Oracle ODAC.Net components. OpsLogix provides all components in a zip file that you can extract and then you run a simple script that install everything you need.

4. After successful installation you will see oracle folder on C:\ with all components:

5. After this procedure you can proceed with importing the OpsLogix Oracle IMP trough the SCOM console

6. You can see that you even have unsealed MP ready for overrides.

7. Everything is imported successful on SCOM 2012 R2.

8. After importing the first thing you need to do is to import your License keys in order to be able to monitor any Oracle database instance.

9. Importing is very easy you have to go to Monitoring -> OpsLogix -> OpsLogix Licensing Dashboard -> Add License.

10. After adding your license you can see all information about the license.

11. Next is to add the node(s) that will act as proxy for monitoring your Oracle database. You go to OpsLogix -> Oracle -> Oracle IMP Configuration Dashboard. The MP automatically detects all your SCOM Windows agents and lists them so you can just select the node(s) that will act as proxy and add them.

12. After adding monitoring node(s) you select Add/Remove instances tab to add databases that you want to monitor.

13. A simple information is added to monitor database as monitoring node, server name, instance name, port and credentials. It is very easy to add databases for monitoring. In fact as I’ve hadn’t any experience with Oracle databases it was harder for to actually configure the databases themselves than to add them for monitoring.

14. Let’s first add DB2 with Oracle 11g installed on it and then DB1 with Oracle 12c.

15. After a couple of minutes you will see both instances being discovered completely. Even that 12c is not supported that instance is discovered with all of its properties:

16. Now let’s see what capabilities this MP has. The MP has Oracle diagram where you can see health map of your Oracle instances and drill down to the root cause of every issue:

17. In the Action menu you will find three tasks:

18. In order to use these tasks you have to copy two files on all of your management servers:

19. You will also find a backup state view. This is very useful for database administrators.

20. You will also find a separate folder for Storage that you will find the different components of the Oracle database architecture in separate views. This helps locating problems faster.

21. What I like most is the Instance performance view where you can see and compare all performance data that is gathered.

22. As we know there is no perfect MP you could always find an error or performance counter that is not monitored for your particular case. The OpsLogix Oracle IMP covers that scenario by providing 3 templates for rules and monitors (Oracle Alert Rule, Oracle Performance Collection Rule and Oracle Two-State Monitor Template) so you can extend the MP very easy.

23. The last I want cover reporting. A lot of good MPs just do not have reporting but this one has one report which is very good. The MP gives you one general report that provides you with availability and performance which are the most common requirements for reports:

As a summary I can say that OpsLogix Oracle IMP is very easy to implement, will offer you with everything you need to monitor your Oracle database and it is definitely SCOM 2012 R2 ready.

SQL Databases Properties are not Discovered by SQL Server Management Pack

I had this issue recently in two separate environments where properties for some SQL databases were not discovered by SQL Server Management Pack as you can see below:

Also collecting performance data for free space of the databases was returning zero (0):

Meanwhile in the Operations Manager log on the SQL servers I was only seeing this error:

Management Group: <MGMTGROUPNAME>.Script: DiscoverSQL2012FileGroups.js : Script ‘DiscoverSQL2012FileGroups.js’ failed.
Inner exception:
Error Number      :
Error Code        : 0
Win32 Facility    : 0
Error Description :
Call stack:Exception.constructor(Script ‘DiscoverSQL2012FileGroups.js’ failed.,Can’t close connection.
Inner exception:
Error Number      : -2146824584
Error Code        : 3704
Win32 Facility    : 10
Error Description : Operation is not allowed when the object is closed.
Call stack:Exception.constructor(Can’t close connection.,Operation is not allowed when the object is closed.
Error Number      : -2146824584
Error Code        : 3704
Win32 Facility    : 10
Error Description : Operation is not allowed when the object is closed.
), ADODB.Close),
Main({D6B26EFE-E183-24E9-DD23-F165CB716A28},{ADFFD930-BDE9-66A1-20F3-CB6FBA80CBC9},<SERVER FQDN>,<SERVER NETBIOS NAME>,MSSQLSERVER),
),
Main({D6B26EFE-E183-24E9-DD23-F165CB716A28},{ADFFD930-BDE9-66A1-20F3-CB6FBA80CBC9},<SERVER FQDN>,<SERVER NETBIOS NAME>,MSSQLSERVER),

So this was my issue but I couldn’t find where exactly was the problem. Some of the properties were discovered and there was even databases that all properties were discovered but performance data for free space was not working on all databases. This lead me think it was some kind of permissions issue but I’ve made the SCOM Action Account “SA” on all SQL servers.

How I fixed it:

1. I’ve created Run As account in SCOM.

2. Named it SQL with type Windows.

3. For credentials I’ve entered the SCOM Action Account.

4. Secured the account only to the SQL servers in my environment.

5. Added this account to these three profiles: SQL Server Default Action Account, SQL Server Discovery Account and SQL Server Monitoring Account.

After these steps all was working as it should be:

So it seems you really need to create Run As account and add it to the SQL Profiles even though the SQL Server Management Pack guide says otherwise:

By default, all discoveries, monitors, and tasks defined in the SQL Server management packs default to using the accounts defined in the “Default Action Account” Run As profile. If the default action account for a given system does not have the necessary permissions to discover or monitor the instance of SQL Server, then those systems can be bound to more specific credentials in the SQL Server Run As profiles, which do have access.

Keep in mind that I haven’t changed any permissions to my SCOM Action Account. This account had all SQL permissions all the time.

My Configuration was:

Windows Server 2012 with latest updates;
SCOM 2012 SP1 UR2
SQL Server Management Pack 6.3.173.1
SQL Server 2012 SP1 CU3

New MP: Microsoft Exchange Server 2013 Management Pack

Exchange 2012 Management Pack was just released by Microsoft. It seems that this is completely new management pack so goodbye to the correlation engine Smile . I’ve never liked that this management pack was some kind of different from all other Microsoft Management Packs. I hope I can test it really soon. You can test it by yourself by downloading it from here. You will not be able to download the guide for this MP as it is located in TechNet Library so here are the links:

New MP: Windows Azure Fabric – Preview

This is a new management pack for Window Azure that can monitor the Fabric. It is replacement MP for the pervious version of Windows Azure MP. It is still in preview so it is not supported in production. Here is overview from download link:

The Management Pack for Windows Azure Fabric enables you to monitor the availability and performance of Azure fabric resources that are running on Windows Azure. The management pack runs on a specified proxy agent and then uses various Windows Azure APIs to remotely discover and collect instrumentation information about a specified Windows Azure resource, such as a Cloud Service, Storage or Virtual Machine. The Management Pack for Windows Azure Fabric provides no functionality on import. For each Windows Azure subscription that contains Azure resources you want to monitor, you must configure discovery and monitoring by first using the Windows Azure wizard in the administration section of the OM Console, then the Windows Azure Monitoring template in the authoring section of the OM Console.

Spread the word:

Like this:

Spread the word:

Like this:

Spread the word:

Like this:

Spread the word:

Like this:

Spread the word:

Like this: