Friday, April 27, 2012

SCOM/OM12 Agent with UAG/TMG ISATAP and IPv6. A special combination it is…

Bumped today into this issue. A customer of mine has a UAG server in place with some basic TMG functionality as well. For the UAG functionality ISATAP is in place as well on the same server. And this can be a challenge in order to get the SCOM/OM Agent to communicate with the SCOM R2 MG.

The Challenge
Since I have already installed the SCOM/OM12 Agent many times on UAG/TMG servers and configured TMG accordingly I thought this would be an easy one. The installation of the Agent went fine. But no way the Agent wanted to communicate with the SCOM R2 MG. Even though I had adjusted the TMG component as stated in the TMG MP document:
image

And yes, I applied the modification as well and checked to see the configuration had been processed properly. But still no luck…

The SCOM event log showed EventID 21006: ‘…The OpsMgr Connector could not connect to <FQDN MANAGEMENT SERVER>:5723.  The error code is 10013L(An attempt was made to access a socket in a way forbidden by its access permissions.).  Please verify there is network connectivity, the server is running and has registered it's listening port, and there are no firewalls blocking traffic to the destination….’.

This really puzzled me. So I started a query in the TMG Console and this is what I got:
image

The Cause
Notice the yellow highlighted IPv6 prefixes. Right after those prefixes the IPv4 addresses are shown. So instead of IPv4 addresses only, the TMG component received the translated addresses instead. And TMG can’t handle IPv6 so it drops them…

When I pinged the SCOM Management Servers from the UAG/TMG server, all I got was those IPv4 to IPv6 translated addresses instead. And vice versa.

The Solution
There are multiple solutions for this one. However, one doesn’t want to break any other functionality as well (like breaking ISATAP for instance). So disabling IPv6 on that server isn’t going to work. Nor disabling IPv6 on the SCOM Management Servers.

A colleague of mine came to this idea, which works like a charm:

  1. On all SCOM Management Servers, edit the HOSTS file so it contains the FQDN of the UAG/TMG server with its IPv4 address;
  2. Flush the DNS cache on those servers after having saved the modified HOSTS file;
  3. On the UAG/TMG server, modify the HOSTS file as well by adding the FQDNs of the SCOM Management Servers and their related IPv4 addresses. Save the file and flush the DNS cache on that server as well;
  4. Restart the Agent service on the UAG/TMG server and BINGO EventID 20070 was shown now, telling me that I had to approve the server in the SCOM R2 Console :).

Now all is well. UAG/TMG and ISATAP are fully functional AND the server is monitored by SCOM….

Many thanks to my colleagues who helped me to crack this one. Thanks guys!

Thursday, April 26, 2012

HP Storage MP v 2.1: The story continues…

It turned out that my previous posting about the HP Storage MP stirred some emotions. Got quite a few comments, most of them were accepted by me and are attached to the same posting. Some of them I deleted since they were a bit too emotional if you get my drift :).

However all comments concur on these issues with the HP Storage MP:

  1. Everything is put into a single MP
    No matter when you only run a 3PAR SAN or a P4500, you get to see EVERYTHING in your SCOM Console. Perhaps HP Marketing thinks this will boost sales (Oh, look a 3PAR SAN! Even though we have a P4500 SAN now, let’s buy an additional 3PAR SAN as well!).

  2. Discoveries running every 60 SECONDS
    This is really sad. Many Discoveries run every 60 seconds! Perhaps HP Marketing is at play here as well. In their ideal world customers add multiple SANs per day…

  3. Almost everything for monitoring is switched ON by default
    The MP itself doesn’t have that many Monitors. But Rules are really present and most of them are enabled by default. Of course, a Rule/Monitor will only run when the Discovered Object is in place. But still it isn’t as it should be.

  4. Many PowerShell instances are run simultaneously on the RMS (up to 30 or more)
    This is bad for performance of the RMS. Noticed the same behavior on the RMS of a customer of mine. Haven’t found the cause for this one yet, still investigating.

Even though I am anything but a MP author, I know to differentiate between a good MP and a bad one. And the HP Storage MP is really awful! Therefore I have exported the HP Storage MP to XML format and edited it in the MP Authoring Console in order to keep only the stuff which is important for my customer who runs a P4x00 type of SAN.

Since I don’t want to get in troubles with HP (for changing their IP and sharing it) I won’t share this MP with the community. But this is what I did:

  1. Exported the MP to XML-format;
  2. Edited the MP in the MP Authoring Console;
  3. Removed the Views which aren’t related to the P4x00 type of SAN;
  4. Disabled all Discoveries which aren’t related to the P4x00 type of SAN nor targeted at SNMP;
  5. Disabled all Rules and Monitors which aren’t related to the P4x00 type of SAN nor targeted at SNMP;
  6. Changed the version number and added additional comments in the Description field;
  7. Saved the results, tested it a test environment;
  8. Sealed the MP;
  9. Deleted the unsealed version of the MP and imported the sealed version.

These are the results:

The Views BEFORE the modifications:
image

The Views AFTER the modifications:
image

The Rules and their status BEFORE the modifications:
image

The Rules and their status AFTER the modifications:
image

The modified MP and it’s description:
image

Since I didn’t want to break the functionality of the MP I didn’t delete any Discovery, Monitor, Rule, Class or whatever. Just tried to keep the MP in tact as much as possible. Next week I’ll be at the customers location and import the MP.

Tuesday, April 24, 2012

HP Storage MP v 2.1: Empty HP Storage P4000 Diagram View is by ‘design’…

Imported the HP Storage MP v 2.1 for a customer of mine in order to monitor a HP P4500 SAN. Followed the guide the whole way through and configured the MP and the other components as required. The test SNMP traps came through so everything seemed to be OK. 

However, the HP Storage P4000 Diagram View stays empty, no matter what I do or try. Also the advised trouble shooting steps (as stated on page 108 of the manual) don’t make any difference. Contacted a colleague of mine who noticed the same behavior in two other SCOM R2 MGs as well…
image

So time to use the other Bing, and soon I found this recent posting, stating:
image

Even though it’s an annoying bug, the relevant Alerts do come in. So hopefully HP will soon release a newer version of the HP Storage MP where this issue is fixed.

Oh, and while HP is fixing this issue, they also should take care of this (as stated on page 84 of the same manual):
image
This is REALLY stupid. Like buying a car but being told it can do anything but drive

System Center 2012 & Veeam: A FANTASTIC Offer!

Veeam celebrates the official release of the Microsoft System Center 2012 suite by offering a free 10-socket license of the Veeam Management Pack for deep VMware monitoring in Microsoft System Center 2012 (OM12 that is of course).

Am I eligible?
You’re eligible when you meet these two requirements:

  1. You’re new to the Veeam MP;
  2. You have System Center 2012 in place or plans to deploy it soon.

Why are you so happy about this?
Good question! It’s just that I love this product since it brings the best out of any OM12 deployment where monitoring of the VMware environment is required. Here are some of best USPs (Unique Selling Points) of any Veeam MP usage, whether you use SCOM R2 or OM12:

  1. VMware Expert out of the Box
    An Alert is not just an Alert. Every Alert contains a wealth of information about the possible causes of the Alert and how to fix them. Many times an Alert contains url’s referring to KB articles of VMware. This way the SCOM Operators don’t require deep knowledge of VMware.

  2. BoBW (Best of Both Worlds)
    Veeam MP integrates seamlessly with vCenter AND OpsMgr without violating anything. It connects to vCenter through valid mechanisms for vCenter and by using a READ-ONLY account. So the Veeam MP can only look into vCenter but not touch anything. By using a proxy-like feature on a dedicated virtual server (VEM, Virtual Enterprise Management Server), the collected information from vCenter is translated to ‘SCOM/OM12’ language. This information is fed to the SCOM/OM12 Agent running on the same server and piped into SCOM/OM12. So for SCOM/OM12 valid mechanisms are used as well.

  3. Use the power of vCenter and SCOM/OM12 to the fullest
    Yes, some basic monitoring can be done by vCenter. But vCenter is by design a resource management tool and nothing more. Where as SCOM/OM12 is a pure bred MONITORING tool. The Veeam MP uses both products for what’s meant for, thus enriching the management of the total VMware infrastructure.

  4. Good Support
    Even when a product delivers value for money, which the Veeam MP certainly does, good support is also a requirement. Veeam delivers awesome support. A shiny example: a key customer of mine had really deep technical questions about the MP while they were only running the test version which has full functionality but only runs for a limited time.

    For them it was crucial to get good answers. So Veeam Support was contacted. However, they couldn’t answer those questions since those questions were really special and more about VMware itself than it related to the Veeam MP. So the support department arranged an online meeting with the developers of the Veeam MP for the next day. The VMware experts of my key customer talked with them for an hour or two and were totally satisfied. They got their answers AND on top of it, additional advice for fine tuning their VMware environment even more. They were very impressed of the deep insights and knowledge about VMware. As a result they wrote an internal business case for buying that MP. And they got it! They tell me the MP has shown them many issues before they became critical, thus making the VMware environment even more available to the business then before the MP was in place.

  5. Good release cycles
    When VMware releases a new version, Veeam will follow soon with a fully supported version also supporting upgrades.

Interested? Go here and be surprised just like me.
image

Friday, April 20, 2012

MMS 2012, an overview

Wow! MMS 2012 was really something special. Especially this year with the release of all new System Center 2012 products, all brought together into a single package, labeled Private Cloud.

I attended many interesting sessions and met - between the sessions – many interesting people. Also I was surprised how many people recognized me because of my blog and complimented me with it. Never thought that my blog carried so much weight. So thank you everybody for your compliments and nice words. Made me feel kind of special.

During MMS an awesome thing happened to me. I have presented a BOF session!!!

How it happened…
Fellow MVP Cameron Fuller was scheduled for presenting two BOF sessions, all about System Center 2012. These two sessions were planned one after another which is a challenge to do. So many fellow MVPs were asked to join the sessions and to help Cameron out.

So I went to those sessions and soon I was given a microphone as well, just to assist Cameron. Soon the first session came to an end and then Cameron walked up to me, introduced me and set me on stage in order to run the second BOF session. He self went away for some minutes in order to get a drink…

Wow! For the first five to ten minutes I was a bit nervous but then it all became better and I started to enjoy it very much. Good questions were asked since the audience knew their stuff as well. Other fellow MVPs were present as well and helped me out when needed. Cameron also returned and assisted me as well.
image

It was an awesome experience which Cameron made possible. So thank you Cameron and many thanks to the audience as well for their good questions. A true community spirit was present those two hours.

Takeaways
These are some of the takeaways of MMS 2012, based on the sessions I visited and the people I spoke.

  1. It’s a stack now, labeled Private Cloud
    Yes, I know. All the System Center 2012 products are like islands, all separate products. BUT… the true power lies within their ability to connect and operate in concert. Only then one has a Private Cloud as Microsoft sees it. These products are (besides Hyper-V of course or any other mainstream hyper visor) crucial to it:

    App Control
    A portal to connect to the private and public clouds, enabling the users to move applications to and from the cloud. Also used for service offerings.

    SCVMM
    The tooling for carving up resources and offering them to the (internal) customers as a private cloud.

    SCOM
    The tooling for monitoring every detail of the Private Cloud.
     
    SCOrch
    The ducktape and mechanism to automate many procedures. The only way to make the cloud self sustaining.
     
    DPM
    For disaster recovery and business continuity.

    SCSM
    Last but not least. The UMBRELLA of the whole cloud solution mainly based on all the Connectors (AD, SCOM, SCCM, SCVMM and SCOrch) so all the CIs contained in the CMDB are really enriched with tons of relevant information. Also the starting point for many automated workflows in conjunction with SCOrch, SCVMM and SCOM.

    In conjunction with the  ‘Cloud Services Provider Pack’  Solution Accelerator, containing templates for SCSM, runbooks for SCOrch and a MP for SCOM, charge back and cost centers are added to the mix, thus enabling a real Private Cloud solution. 

  2. MP Authoring tools are growing up
    Finally, the tools for MP authoring have grown up. No more awful MP Authoring Console but two new tools, each of them targeted at a different audience:

    IT PROs
    For IT PROs Visio MP Designer is build. This tool uses Visio 2010 PREMIUM and enables on to create good MPs in a fast and highly visualized manner. When one has a LOB which needs customized monitoring this tool is really nice to use. One can visualize the LOB in Visio and create the required MP in no time!

    Developers
    For developers Visio MP Designer won’t do at all. For them Visual Studio has a new add-on, enabling one to create MPs like a real developer, with all the control mechanisms in place and IntelliSense.

    These tools are about to released to the public this week.

    See AM-B311 and FI-B313.

  3. APM (Application Performance Monitoring)
    In OM12 APM is something really special. It’s the latest version of AVIcode and fully integrated into OM12. It enables organizations to look inside .NET based applications in order to catch exceptions and get rich information about it. Pete Zerger and Alex Fedotyev gave an awesome session about it. Learned a lot from it.

    See AM-B310.

  4. OM & SCOrch, a happy marriage
    How two totally different products are capable of enhancing one another thus enabling better monitoring and remediation. Brian Wren gave some demonstrations which are very usable in everyday practice of any IT shop.

    See DB-B305.

  5. Australia and UK together: How one plus one makes three!
    There was this session all about advanced analytical reporting, presented by Gordon McKenna (UK) and Sean Roberts (Australia). Besides a lot of humor (and strange voices) they shared good information and insights about how to get the most out of your SC 2012 implementation when reports or visual representation of crucial information is required. A very good session it was. And refreshing as well because they really had fun while presenting.

    See FI-B326.

  6. Service Delivery and automation
    All about the Private Cloud and the importance of Services. How to create them, offer them and automate much of the underlying processes. Very informative session it was.

    See SD-B308.

  7. Microsoft has embraced JAVA?
    No, not really. Phew! But… OM12 is very capable of monitoring JAVA applications. In a session this was clearly demonstrated and explained. I don’t do Java (my devices and websites do but no interaction of myself is required here) but even I understood the mechanisms of it all. A good session it was.

    See AM-B318.

  8. Microsoft knows how to PARTY!!!!
    Yesterday evening Microsoft invited all attendees to a party. And boy, did this party ROCK! Awesome! As location the casino/hotel Paris was chosen. Under the Eifel Tower (!) – set in thousands of lights – and at the pool deck the party was pumping. An awesome surrounding. There was plenty of food, served by beautiful women (loving Vegas!) and drinks. There was a good DJ and a drummer, really making the party go. An awesome experience I’ll never forget. Thank you Microsoft!!!

    Some impressions:
    image

    Aq7W8kNCIAEkEv1

    image

    image

    image

So it was an awesome experience. Combined with Las Vegas even a bit crazy. Had the time of my life.

When you haven’t been at MMS 2012, you still can watch online the sessions. Go here, sign up (for free!) and check out the sessions. When you’re interested in the sessions I have mentioned in the posting, search for the related session number as listed per topic in this posting.

Wednesday, April 18, 2012

Microsoft System Center 2012 is RTM!

For some time now I was under strict NDA and therefore not allowed to blog about it. Other bloggers weren’t under NDA and noticed it some time ago already: Microsoft had released – silently  - the RTM bits of all System Center 2012 products, and told about it on their blogs.

However, Microsoft decided not to put it into the spotlight but wait until the 17th of April 2012 where Brad Anderson would officially announce the System Center 2012 products to be RTM and GA (General Available):
image
(I followed most of the keynote live at MMS but had to leave early since I had to do exam 70-246. Sorry Brad! :) )

Since it’s the 18th of April, I am finally not restricted anymore to blog about it, so here it comes (what many of us already knew):
Microsoft System Center 2012 – also referred to as Microsoft Private Cloud – is since April the 17th officially RTM and GA!!!!
Much has changed in the 2012 release of the System Center products. Microsoft positions it as ‘…a cloud and datacenter management solution that provides a common toolset to manage private and public cloud applications and services…’.

In order to put their money where their mouth is, they have dramatically changed the licensing model. In the ‘old’ days – for the previous System Center products – there were over 200 different SKUs. Now for System Center 2012 there are only TWO SKUs where a single license entitles you to install ANY System Center 2012 product with SQL Server 2008 R2 Standard SP1 included for free! Go here if you want to know more about the details.

Say what? MCSE certification is BACK?! Uh… Yes and NO!
On top of it, Microsoft has changed the certifications as well. Now one can’t do separate exams for the System Center 2012 products anymore. Instead, most of the products are combined into two separate exams: Exam 70-246, Monitoring and Operating a Private Cloud with System Center 2012, and Exam 70-247, Configuring and Deploying a Private Cloud with System Center 2012.

In conjunction with the exam which certifies one as a MCSA Windows Server 2008, it creates the certification MCSE Private Cloud. In this case however MCSE means: Microsoft Certified Solutions Expert. This website tells you more about it.

Mind you however, exams 70-246 and 70-247 are still in beta. While attending MMS 2012 one can do these beta exams for free. I did both and boy, these exams are really tough. Because now, all the products are related to each other. So only a good understanding of OM12 or any other System Center 2012 product won’t suffice anymore.

No way! One has to know every System Center 2012 product AND how they relate to each other. For me personally Exam 70-246 was doable. But exam 70-247 really filleted me. I feared the Next button since many questions were mumbo jumbo to me. So there is a lot of new grounds for me to be covered in order to pass for this exam.

Since these exams are beta, one doesn’t get to see the end result (failed or passed) at the end of the exam. After a few weeks (6 to 8 that is…) I’ll be notified. Exciting times…

MVA? Never heard about it?
In order to get to grips with the new technologies and the way Microsoft positions them, one needs to gain a lot of knowledge and experience as well. Gladly there are already very good resources to be found like:
  1. MVA - Microsoft Virtual Academy
    MVA is an online (and FREE!) collection of resources where one can train and learn all about the latest Microsoft Technologies at your own pace. 
    image
    This is an awesome source of much knowledge and real life experience. There are tracks all about the Private Cloud and the related System Center 2012 products. Every track consists out of multiple parts. Every part consists out of a comprehensive video and the related transcript in pdf-format. Every part tests your knowledge by asking some questions in multiple choice format. Only when enough questions are answered correctly, one can start the next part of that track.

    Again, MVA is free and can be found here. I strongly advice you to go there and follow the tracks related to your job as is today and the tracks required for job as it will be tomorrow. Seriously, in our trade one needs to plan ahead and stay up to date.   
         
  2. Private Cloud Jumpstarts
    A nine part video series all about the Private Cloud (Microsoft System Center 2012 products) how to deploy and operate them.

  3. TechNet Virtual Labs
    Without a doubt there are more to be found, but this one I know. TechNet Virtual Lab: System Center 2012 Operations Manager: Infrastructure and Application Performance Monitoring.
Also new books will come out, all about preparing for the new certification and the related exams.

Wrap up and some words of advice
As you can see, these are exciting times. New products, new positioning, new certifications and new exams. Our trade isn’t boring at all and introduces many new challenges. The journey to the (Private) Cloud has started, whether we like it or not, it’s up to you to board that train. When you don’t or fail to be there on time, you’ll have a fair change to find an empty platform and a deserted train station as well.

Our trade is changing fast and we need to reinvent ourselves. When we don’t we’ll most certainly become obsolete and find ourselves replaced by very advanced tooling. In order to stay on top of things we need to master the skills required to master the very set of tools which could throw ourselves out of business. This way we have prolonged our future in the IT.

Monday, April 16, 2012

MMS 2012: Las Vegas Baby!

Wow!

In 2010 some colleagues and me had planned to go to MMS 2010, also in Las Vegas. Everything was set to go but then an ash cloud came along and threshed our plans. In 2011 other things were at play, so no MMS 2011 for us either. But now in 2012 we’re at MMS!

Arrived last Saturday evening and yesterday we went out to experience Las Vegas. If there is one word to describe Las Vegas itself, it’s BIZAR. Really amazing to see a whole city dedicated to entertainment in many extravagant shapes. Entertainment is reinvented here and brought to new ‘heights’.

Competition among all the casino owners – in order to lure the people into their casino’s and to make them stay away from their competitors – has introduced a whole new level of eye ‘candy’ and ‘shock and awe’, unprecedented any where else but Las Vegas.

Some impressions:
image
The Eiffel Tower in Las Vegas? Yes, a small version, but still pretty high!

image
This is not really sky but the painted ceiling inside the Venetian Hotel. It represents the St Marco Square of Venice, including the buildings AND the channels WITH the gondola’s. The street isn’t wet, but it’s some kind of special coating.

image
The hotel where we’re staying. Even with hundreds of rooms, this hotel is relatively small compared to the Venetian which has thousands of rooms.

image
Malls inside the hotels which are way really huge. They stretch as far the eye can see and consist out of multiple levels. Everything inside the hotels! Never seen such a thing before. The seize and again, the impressions, are awesome!

However, this blog isn’t a travel guide. So I’ll leave it at that. I’ll hope to post soon about MMS 2012 on this blog, so stay tuned.

Friday, April 13, 2012

SCOM R2 to OM12 upgrade: A caveat when upgrading SQL 2005 to SQL 2008 R2 SP1

When one has a SCOM R2 environment in place based on SQL Server 2005 and want to upgrade to OM12, it’s required to upgrade to SQL Server 2008 R2 SP1 first.

Even though I have posted a whole series of articles about that upgrade process, there is one small but important detail to reckon with. Otherwise OM12 will generate errors in the OpsMgr event log of the OM12 Management Servers.

It’s all about the OpsMgr and OpsMgrDW databases which still run in the SQL 2005 compatibility mode. OM12 doesn’t like that and will throw errors in the OpsMgr event log.

KB2700028 tells you how to solve it.

OM12 Design Guide

The OM12 Team has released a new guide, all about designing a OM12 environment. Since the guide is small (only 10 pages of information) it’s easy to read while it highlights the most significant changes compared to SCOM R2.
image

So this guide is an excellent stepping stone to designing and implementing your OM12 environment (or upgrading your SCOM R2 environment) without loosing yourself in too many documents or too much information.

Want to know more? Go here.

Thursday, April 12, 2012

How To Fix: Not all Exchange 2010 Mailbox servers are discovered by the Exchange 2010 MP

Sometimes when the Exchange 2010 MP is imported not all Exchange Mailbox 2010 servers are discovered, no matter what you try, like flushing the cache of the SCOM R2 Agent.

This posting is all about this issue, its cause and how to remedy it.

Issue
The Exchange 2010 MP is imported and configured but not all Exchange Mailbox servers are discovered. They run a SCOM R2 Agent none the less and these Agents work properly. All other MPs land properly (Server OS for instance) and run as expected. Even when the Agent cache is recycled, the issue is still there.

Cause
Many times Exchange Mailbox Servers having these issues are still under construction, or better, the installation isn’t finished. When the ExBPA (Exchange Best Practices Analyzer) is run against servers like those, this is shown:
image

Also when the registry on servers like those is checked, this is shown:
image

Remedy
This posting of the Exchange Server Team is all about similar issues with Exchange 2007 Servers.

Restart the Exchange Setup. This will finish the configuration. When checking the registry again, the highlighted regkey (MailboxRole) will only contain two values: UnpackedVersion and ConfiguredVersion, both with the same value.

Now the Exchange 2010 MP will discover this Exchange Mailbox server as intended.

All credits go to Maarten Piederiet who pointed this one out to me. Thanks Maarten!

BizTalk Server 2010 MP: Let’s make ‘some’ noise…

Anyone who uses (or used…) the BizTalk Server 2010 MP knows this MP creates TONS of noise. Which is bad since it makes this MP useless.

So many organizations implemented the MP and removed it soon afterwards. Gladly this MP has been modified by a well respected fellow MVP, Bob Cornelissen. And true to the spirit of the Community, shares this modified MP with us all!

So for anyone interested in monitoring BizTalk Server 2010 but removed the original MP since it creates tons of Alerts, go here and download the modified MP.

Please take note: The original BizTalk Server 2010 MP is required for the modified MP to work!

All credits go to Bob Cornelissen of course. Thanks Bob for your efforts and keeping up the true Community spirit!

Tuesday, April 10, 2012

New KB: How to add a product key to an eval version of OM12

Microsoft has released a new KB article about how to upgrade an eval version of OM12 to a fully licensed version of OM12. For customers participating in TAP programs this might be valuable information.

KB2699998 describes this process in more detail.

Migrating from SCOM R2 CU#5 to OM12. Part II: Prepare yourself…

----------------------------------------------------------------------------------
Postings in the same series:
Part I   - The environment
Part IIIPre Upgrade Tasks for Operations Manager
Part IV – Upgrade Tasks for Operations Manager
Part V  – Upgrade Tasks for Operations Manager, continued
Part VI – Wrapping Up 
----------------------------------------------------------------------------------

In the second posting of this series I’ll describe the preparations which need to take place prior you start the actual migration. Please take note that during the migration itself other preparations are required as well, also referred to as Pre-Upgrade Tasks (two sets of them actually) and one set of Post-Upgrade Tasks.

Before I start however I want you to be aware of these two important items (A & B):

A: Dead links
The set of (online/offline) OM12 documentation Microsoft offers isn’t totally error free unfortunately, like containing dead links. For instance, the links in the (online/offline) documentation referring to the System Center 2012 - Operations Manager Upgrade Process Flow Diagrams don’t work many times and result in a Page Not Found error:
image

Or you end up on a page which only shows a part of the diagram and isn’t clickable at all..

Also some links referring to the Supported Configurations page results in the same Page Not Found error. So for the completeness I put the set of working links (at this moment that is) here:

You’ll find yourself many times on these pages since they contains tons of worthwhile information.

B: What SQL Collation settings do I need?
The (online/offline) documentation isn’t very clear here. For instance based on the information on this page under the header Verify the SQL Collation, one might think that different collation settings are supported by OM12.

However, IMO(!), the only real supported collation setting is SQL_Latin1_General_CP1_CI_AS as stated on this website all about the Supported Configurations for System Center 2012 - Operations Manager. As far as I am concerned, that page is leading. So make sure you run that SQL Collation setting.

Some MPs (Exchange 2010 Reporting MP and the nWorks Veeam MP) require that SQL Collation setting in order to function properly. So be warned!

OK, now it’s time to start.

Complex or Simple Upgrade Scenario
Basically Microsoft considers two kinds of upgrade scenario’s. Simple or Complex. By no means a Simple migration doesn’t refer to a small scaled SCOM R2 environment. Nor does a Complex migration refer to enterprise scaled (distributed) SCOM R2 environments.

Simple or Complex refers in this case whether your SCOM R2 environment runs on by OM12 supported software. The most important components here are the Server OS version for your Management Servers (RMS, MS and Gateways) and SQL server and the version of SQL Server:

  1. Required Server OS for RMS, MS and Gateway Server:
    Windows Server 2008 R2 SP1;

  2. Required Server OS for SQL:
    This might seem a no-brainer but there is caveat here. For SCOM Reporting Windows Server 2008 R2 SP1 is required where as for the other SQL Server roles (Operational database and the Data Warehouse) Windows Server 2008 SP2 (or higher) will suffice.
    In smaller environments a single SQL Server runs both databases and hosts the SSRS instance for SCOM R2 Reporting as well. For this kind of SQL Servers Windows Server 2008 R2 SP1 is required! So check and double check!

  3. Required SQL Server version:
    At least this is the same SQL Server version across all OM12 roles (Operational database, Data Warehouse and SSRS): SQL Server 2008 SP1 or higher (SQL Server 2008 R2, or SQL Server 2008 R2 SP1).

When these components are supported by OM12 you have a Simple upgrade scenario and when not, well you find yourself in a Complex upgrade scenario. Of course, when your SCOM R2 environment runs on non-supported OM12 Server OS and/or SQL Server version, you’ve got ‘some’ work to do BEFORE you can even think of migrating SCOM R2 to OM12. However, Microsoft has documented these steps as well.

But even when your SCOM R2 environment runs on SQL Server 2008 SP1 or higher and the RMS, MS servers, Gateway Servers and SQL Server(s) run on Windows Server 2008 R2 SP1, the migration scenario can still become a complex one? Why? Read on…

Complex and Simple Upgrade Scenario’s in Real Life
IMHO, there aren’t many SCOM R2 environments which run only Microsoft software. I have seen many SCOM R2 environments which run third party software like, but not limited to:

  1. Third Party MPs (Veeam, OpsLogix, Jalasoft, Bridgeways, NiCe etc etc);
  2. Third Party add-ons (Savision Live Maps™, Savision Vital Signs™, etc etc);
  3. ACS add-on Secure Vantage (SVT).

For this kind of software, contact your vendor in order to find out whether this software (or better the version) supports OM12. And not just that, also check whether new or updated licenses are required.

Sometimes third party software can be a challenge to migrate to OM12. For instance SVT might require a whole upgrade scenario of its own when you’re running an old version of SVT, before you can even consider the migration to OM12.

Another thing to reckon with are the MPs which are supplied by the community, like (but not limited to):

  • xSNMP MP Suite
    When this MP is in place, your SCOM R2 environment can be upgraded to OM12. The xSNMP Suite will continue to work but you can’t add new SNMP devices since the base discovery class is changed in OM12 nor can you rediscover these devices. Which basically means the xSNMP MP becomes static in your OM12 environment. (Thanks to Pete Zerger for this information);

  • OpsLogix Ping MP
    This MP adds some functionality to the regular SCOM R2 Console by using a special dll file. OpsLogix has stated they’re working on an OM12 version. The dll used for the SCOM R2 Console won’t work with the OM12 Console. So remove that file. Your configuration for the OpsLogix Ping MP will stay in tact as long as the OpsLogix Connector isn’t removed.

  • Addendum MPs
    For some MPs there are addendum MPs (Server OS and AD for example). Check these MPs before migrating a SCOM R2 environment which runs those MPs to OM12.

  • SCC Health Check Reports MP
    This MP seems to work properly in OM12.

  • Logical Disk Space Report
    No issues. Works fine in OM12.

  • PKI Verification MP
    This MP seems to work properly in OM12.

Therefore, build yourself an OM12 test environment and test those MPs in order to see whether all is still OK. In some cases it might be required to remove some MPs before starting the migration. Or, remove some parts of it, like the dll file used for the OpsLogix Ping MP since the SCOM R2 version isn’t compatible with the OM12 Console.

As you can see, even when your SCOM R2 environment runs on supported Server OS versions for OM12 and SQL is OK as well, there still might be some challenges which need to be addressed. And: NEVER ASSUME! Always be sure! So build that OM12 test environment.

So the preparation for migrating to OM12 comes down to these steps:

  1. RTFM the official Microsoft documentation about the upgrade process. Not just once but a few times. And don’t stop thinking. When in doubt go the forums and post your question;
  2. Familiarize yourself with the Upgrade Process Flow Diagram which fits your situation. Know what steps you have to take and in what order.
  3. Check whether the Server OS and SQL Server versions are in line with OM12. When not, fix it. Might sound easy but it isn’t, I know from first hand experience. So take your time;
  4. Check the third party MPs and add-ons you’re running. Check whether they’re OM12 compatible. And check the licensing requirements. When something isn’t OM12 ready, fix it according the  advices as set out by the vendor;
  5. Know what you’ve got, like (but not limited to):
    1. RMS;
    2. How many MS Servers;
    3. How many Gateway Servers;
    4. How many SQL servers are being used for SCOM R2;
    5. Which server hosts the SCOM Web Console;
    6. How many pushed SCOM R2 Agents;
    7. How many manually installed SCOM R2 Agents;
    8. The SCOM R2 service account credentials.
  6. Build yourself a test environment, even when it’s only a single-server solution (it’s better to run two OM12 MS servers, but when resources are really limited for some basic testing, a single-server solution will suffice);
  7. Check the backups of the RMS, MS servers, Gateway Servers and SQL Server(s) used for SCOM R2. Are they in place, functional and OK?

Only now you’re ready to start the upgrade as described in the documentation provided by Microsoft. In the next posting of this series I’ll describe the first steps in the upgrade process, so stay tuned.

MMS: Travel- & Survival Guide and a list of useful resources

Dieter Wijckmans, a Belgium based System Center specialist has posted a whole series of articles all about surviving MMS 2012. These postings are fun to read and contain some good advices.

Want to know more? Go here: http://scug.be/blogs/dieter/archive/2012/03/06/a-travel-guide-how-to-survive-mms2012.aspx

A list of useful resources all related to MMS can be found on the website of My IT Forum: http://myitforum.com/myitforumwp/microsoft-management-summit/.

New MP: App Controller MP

A few day ago Microsoft released a new MP for monitoring System Center 2012 – App Controller, version 1.0.1200.0.

Taken directly from the website:
image

MP to be downloaded from here.

New MP: Orchestrator MP

A few day ago Microsoft released a new MP for monitoring System Center 2012 – Orchestrator, version 7.0.

Taken directly from the website:
image

MP to be downloaded from here.

Friday, April 6, 2012

Migrating from SCOM R2 CU#5 to OM12. Part I: The environment

----------------------------------------------------------------------------------
Postings in the same series:
Part II   - Prepare yourself
Part IIIPre Upgrade Tasks for Operations Manager
Part IV – Upgrade Tasks for Operations Manager
Part V  – Upgrade Tasks for Operations Manager, continued
Part VI – Wrapping Up 
----------------------------------------------------------------------------------

As promised I will start a new series of blog postings all about migrating from SCOM R2 CU#5 to OM12. This is the first posting in this series.

Especially for this series of postings I have build myself a whole new SCOM R2 environment in my test lab. The first posting in this series will describe this environment in more detail. This way we know what were ‘up against’ when we start the migration. It’s very important to know what you have before starting the migration. Otherwise one might end up in a bad situation where the migration goes wrong and results into a wrecked SCOM/OM12 environment. So let’s start.

What we have
Since resources are limited I have build a small and yet a representative SCOM R2 environment, which resembles (in the basics!) many distributed SCOM R2 environments.

There are two different forests in place, without a trust. In one forest (SCOM2OM12.local), there is a DC, a RMS, a MS and a SQL server in place. Also a PKI is in place and fully configured. In the other forest (NTE.local), there is a DC, a dedicated Gateway Server and a member server in place. The Gateway Server communicates with the MS server using certificates. The servers behind the Gateway Server, communicate with it using Kerberos.

In total 7 servers are running, based on Windows Server 2008 R2 SP1 with the latest patches and updates applied. SQL server is based on SQL Server 2008 R2 SP1 CU#4. SCOM R2 is installed with CU#5 applied. These are the server names and their functionality:

Forest: SCOM2OM12.local (boy, do I have some imagination…)

  1. SCOM-DC01: AD Controller for forest SCOM2OM12 and Enterprise PKI;
  2. SCOM-DB01: SQL Server with SSRS;
  3. SCOM-MS01: RMS;
  4. SCOM-MS02: MS.

Forest: NTE.local

  1. NTE-DC01: AD Controller for forest NTE;
  2. NTE-GW01: SCOM R2 Gateway Server;
  3. NTE-SV01: Member Server.

In a simplified Visio drawing the environment looks like this:
image

What MPs are loaded
All basic Microsoft MPs are in imported and fully configured:

  1. Server OS;
  2. AD;
  3. DHCP (oink!);
  4. DNS;
  5. IIS;
  6. SQL;
  7. Windows Terminal Services;
  8. Windows Remote Desktop Services;
  9. GPO;
  10. DFS Namespaces;
  11. DFS Replication;
  12. SMB;
  13. File Services;
  14. Certificate Services.

On top of it, I have also imported these free MPs, which really add value to any SCOM R2 environment:

  1. OpsLogix Ping;
  2. SCC Health Reports;
  3. PKI Verification MP.

Even though IMO the xSNMP Suite MP rocks, with OM12 this MP isn’t needed any more (I hope). And for another thing, the SNMP module in OM12 is totally rewritten, from the ground up. So the xSNMP isn’t compatible with OM12 anymore. At least, that’s what I expect from it.

On top of it I have created some bogus Distributed Applications, DA 01 and DA 02. Also some wizard based monitors are in place:

  1. Process Monitor. Process HealthService.exe is being monitored;
  2. TCP Port Monitor. Checks port 80 on the RMS;
  3. Web Application Monitor. Checks the availability of the SCOM Web Console;
  4. Windows Service Monitor. Checks the availability and running state of the service MSDTC on the RMS.

Also some additional Views and Roles are created and a SMTP Notification Channel is in place. This way we have a good environment in place for the migration. In the next posting of this series I’ll talk about the upgrade path, the documentation which is available, the preparations and the lot. See you all next time!

Tuesday, April 3, 2012

OM12: WS-Management/SMASH Device Discovery Template Released

Yesterday Microsoft released the WS-Management/SMASH Device Discovery Template for OM12.

Picture taken from this website.

What it is and what is does? Go here and you’ll know everything what you need to know…

Visio and SharePoint extensions for OM12 Released

Yesterday Microsoft released the Visio and SharePoint extensions for System Center 2012 – Operations Manager (OM12). Even though they don’t offer the same functionality as Savision Live Maps™ (and the simplicity), these extensions are still nice to have.
image

Extensions can be downloaded from here. More information about how Microsoft positions these tools, go here.

Updated MP: SQL Server Monitoring MP, version 6.3.173.0

I wanted to download the SQL Server Monitoring MP by using the Catalog through the SCOM R2 Console. However, the SQL MP wasn’t there to be found:
image

That’s strange. Normally when this is happening, there is an update of that MP being pushed out by Microsoft which takes a while to go around the globe. So it’s time to visit the MP Catalog website. And yes, there it is, the NEW SQL Server Monitoring MP:
image

So the SQL Server MP is updated. Nice! Since from now SQL Server 2012 is officially supported (taken directly from the website):
image

Other changes are:

New features:

  • AlwaysOn Monitoring
    • Automatically discover and monitor availability groups, availability replicas, and availability databases for hundreds of computers.
    • Health roll-up from availability database to availability replicas.
    • Detailed knowledge with every critical health state to enable faster resolution to a problem.
  • Seamless integration with Policy based management (PBM)
    • Auto-discover custom PBM polices targeting AlwaysOn and database components.
    • Rollup of health of policy execution within the SQL monitoring pack under extended health.
  • Support for Mirroring and Replication Monitoring (only applicable to SQL Server 2008 and 2008 R2 version of management pack)
    • Discover mirroring databases, witness, and mirroring group.
    • Monitor database mirror state, database mirror witness state, and mirroring partners’ state.
    • Custom diagram view to visually represent the primary and the mirrored databases.
    • Approximately twenty rules to detect replication events.
  • Improved Freespace monitoring with mount point support

So a very good MP even got better! Very good job guys! The new SQL Server Monitoring MP can be downloaded from here.

New KB: Monitoring Text and CSV log files in SCOM

Microsoft has published an updated KB article all about Monitoring Text and CSV log files in SCOM.

Even though this KB article is a bit difficult to read, it contains good information. Ever wanted to monitor log files but don’t know where to start? Read KB2691973 and you are well under way.

Monday, April 2, 2012

Documentation: Technical documentation download for OM12

For a few days now Microsoft has officially released the first version (1.0 !) of the set of technical documentation for System Center 2012 – Operations Manager (OM12):
image

Files can be downloaded from here.

Same sets of technical documentation can be downloaded as well for these System Center 2012 products:

  1. Configuration Manager: http://www.microsoft.com/download/en/details.aspx?id=29264
  2. Virtual Machine Manager: http://www.microsoft.com/download/en/details.aspx?id=6346
  3. Orchestrator: http://www.microsoft.com/download/en/details.aspx?id=29258
  4. Service Manager: http://www.microsoft.com/download/en/details.aspx?id=27850.

OM12 Sizing Helper Tool

Since a few days Microsoft released the new Sizing Helper Tool for OM12. To be found here.
image

How To: Enabling tracing to troubleshoot Notification issues

Jonathan Almquist has written a a good posting about how to enable tracing on the RMS when there are serious issues with sending out Alerts (Notifications).

Want to know more? Go here.

SCOM R2: EventID 2023.

Bumped into an issue where the OpsMgr event log of an OpsMgr Agent logged EventID 2023: ’…The health service has removed some items from the send queue for management group since it exceeded the maximum allowed size of xyz megabytes…’. This event was shown once per hour.

Cause
As it turned out, this server ran many monitored objects, services and applications. As a matter of fact, too many. Normally the queue for a SCOM Agent is set to 15 megabytes (15360 KB).

Solution
The maximum queue size is set to 75 megabytes (76800 KB). Afterwards the Agent is restarted and all is well again. The maximum queue size is 256 megabytes (262144).

The register key you need to update is to be found here: HKLM\SYSTEM\CurrentControlSet\services\HealthService\Parameters\Management Groups\<MG NAME>\MaximumQueueSizeKb.

Advice
Whenever you bump into a server where there are really many objects to be monitored and not all the collected information seems to reach SCOM, try these steps as described here:
image

This applies to SCOM R2 Agents and OM12 Agents as well.

SCOM R2/OM12: Monitoring Oracle

Oracle isn’t just a database solution. There is way much more to it and today many businesses rely on Oracle databases for many business critical solutions. So SCOM needs to monitor those databases as well. 

For a long time there was only ONE solution available for it, delivered by EXC. Later acquired by Quest. However, these EXC MPs are a nightmare. They are really bad so I stay away from solutions like that.

Sometime ago a new player entered the market, Bridgeways. They offer an Oracle MP as well. However, a key customer of mine tested that MP and soon we discovered it has some issues. For instance in the beginning there was no good support for multi listeners, which is the equivalent of a single server running multiple SQL server instances. Later on that support was added, but still there were other issues at hand as well. Like configuring the MP itself and the support we got. IMO, the total experience of a MP consists out of many aspects, among them the support a customer gets.

So this key customer of mine decided to run some synthetic transactions instead and to delete the Oracle MP since it added more work than it added value. Back to square one.

Therefore it’s good news that TWO new MPs for monitoring Oracle will see the light. When? Just contact these two companies and they’ll tell you more:

  1. NiCe Management Solutions: Oracle MP, go here.
  2. OpsLogix: Oracle MP, go here.

Don’t ask me what MP is better. As stated before the TOTAL quality of a MP is based on many aspects, like these ones:

  1. Availability of the MP;
  2. Support on the MP;
    1. When you contact support, how long does it take before you get a valid answer?
  3. Usage:
    1. Does the MP deliver what’s expected from it?
    2. Is the MP easy to handle (import, configuration) or is it rocket science?
    3. Does the MP deliver good reports?
  4. Visualization:
    1. Is the information shown in a proper way in the SCOM Console?
    2. Is  there a good State View?
    3. Is there a good Event View?
    4. Is there a good component overview, enabling good event analyses?
  5. Release cycles:
    1. How many times is the MP updated (for bugs or monitoring new releases of the monitored product/service)?
    2. How RTM is the MP? Does it contain bugs which are bad or annoying?
  6. Pricing.

As you can see, the price is important but not the most important component. A MP which comes cheap but doesn’t deliver any added value or even worse, floods your SCOM environment with false Alerts or wrecks your databases because of too short discovery cycles, broken cook down, or config churn, you pay a HIGH price.

My advice:
Test these MPs against the scenario’s your Oracle DBA’s have set out. You’ll see that for every MP there are some downsides to note. Choose the MP which is the best fit for you.

Updated MP: Active Directory MP, version 6.0.7822.0

A few days ago Microsoft released the updated version of the AD MP, version 6.0.7822.0.

Changes in this update, taken directly from the website:
image

The AD MP is one of the MPs I really like. It’s a good and solid MP which really adds value to your SCOM/OM12 environment. MP can be downloaded from here.

One additional advice:
This MP consists out of many files, like the Microsoft.Windows.Server.AD.ClientMonitoring MP. However, this file is not really required and adds some extra’s to the monitoring of your AD functionality AND availability. Many times, less is more. The same situation applies here. Import all the required components of this MP and forget for a while the file Microsoft.Windows.Server.AD.ClientMonitoring. Configure this MP as required and when all is well and you’re missing out on the monitoring of the AD from the client perspective, import the file Microsoft.Windows.Server.AD.ClientMonitoring and configure it accordingly. Saves one a lot of time and many times, the AD MP without monitoring from the client perspective already adds enough insight about the health state of your AD environment.