Rory's profileSystem Center TrainingPhotosBlogListsMore Tools Help

Blog


    March 21

    Object Discovery Frequencies

    Object Discovery Frequencies… Be Careful

    How often do you set the frequency of your object discovery? A number of the Microsoft object discoveries are set to 3600 seconds or one hour while others are set to 7200 seconds. If you want your object discoveries to run less frequently you could set it to something larger but be careful. Setting your object discovery frequency to greater than 86400 will generate an error.

    object discovery frequency

    The unfortunate part of this error is that you aren’t able to change the frequency in the GUI as this is the error that you receive when you attempt to access the properties of the attribute whose creation generated the object discovery.

    Lesson: Don’t set your frequency above 86400.

     For great training on Operations Manager 2007 and other System Center technologies, check out our offerings at http://www.infrontconsulting.com/events.htm.

     

    Ops Mgr Pros and Cons

    Ops Mgr Pros & Cons

    Every application has it’s challenges and the learning curve is certainly one of them. Are they things that I feel we missed in Ops Mgr; yes definitely but Ops Mgr also provides for more flexibility in how and what you monitor. I can start a list of what I envision as improvements and another of what is a pain in the but.

    Pros for Ops Mgr 2007

    -          More granular notification capabilities. The notification model is now subscription based. The ability to add up to 3 SMTP servers and define failover is included as is support for IM and SMS text messaging.

    -          Maintenance mode. Now you are able to put an individual object into maintenance mode, not only an entire computer.

    -          Single console. You can now do almost everything (except advance rule creation and report writing) through a single console.

    -          Much improved security. The product no longer relies on local computer groups (complete insanity and an administrative nightmare in MOM 2005) but rather leverage role based authentication.

    -          Sealed Management packs prevent force customers to adhere to best practices and save overrides in a separate MP.

    -          New architectural roles. With the addition of the Gateway server role, the ability to monitor systems in an untrusted domain or DMZ is significantly improved. The architecture is much more scalable and easy to extend for growing organizations.

    -          Monitors. The concept of monitors and their ability to auto resolve greatly reduces the noise in the Monitoring node of the Operations Console. This makes the console much easier to work with as an operator.

    -          Improvements to the Web Console. The web console now allows for tasks to be executed, object to be placed in maintenance mode.

    -          PKI support. The support for PKI as an alternative authentication method is a great addition though usually a little difficult to get working at first. This has little to nothing to do with Ops Mgr.

    -          Reporting. Reporting is much, MUCH faster and the Microsoft generic report library allows you to report on any group of objects or individual object. Reports are integrated into the Ops Console and the same security model applies to the reports as to the Ops Console.

    -          Health Explorer. When an alert is generated, an operator can very quickly identify where the alert is through health explorer and see the date and time the change has occurred. You are also able to embed diagnostics and recoveries into monitors and this can allow for an operator to run a response in the context of the health service or simply view diagnostic information about the event that generated the alert. Recoveries can also run automatically and resolve issues based on known resolutions. When an alert is resolved automatically, it’s state is also updated and the alert resolution state changed to closed in the console, reducing the noise in the console.

    -          Connectors into other operations manager products are now free from Microsoft through their purchase of Engyro.

    -          ACS!! A very scalable and efficient security event log collection mechanism has been included in Ops Mgr. This separate service operates over a separate port and normalizes and transmits data very efficiently to the new Ops Mgr collector mgmt server role.

    -          Each MP is broken into 4 categories that make up Microsoft’s health model; Availability, Configuration, Performance and Security.

    -          There have been significant improvements in the ability to monitor line of business applications. It’s critical however that this be planned out as the use of multiple unsealed MPs will get you into trouble. Sealing your MPs and having an MP management lifecycle plan is crucial to your success.

    -          The discovery of network devices is now supported.

    -          Web transaction monitoring has been greatly improved and is now native to the application. This makes the ability to monitor a web site very easy.

    -          The ability to easily configure TCP Port, OLE DB, and service monitors has been included natively through Management Pack templates.

     

    Cons / Issues with Ops Mgr 2007

    -          Complexity. This product has a LOT of hidden or buried features that require you understand the architecture to really leverage. (i.e. targeting of tasks object classes)

    -          Maintenance mode. There should be a GUI to schedule maintenance mode. This can be done through a Powershell script. Putting an entire computer into maintenance mode is more complicated and requires that you put three different objects into maintenance mode.

    -          Security. The ability to create a report user role is missing in the Ops Console. This can be done through the Command Shell but should be included in the Ops Console.

    -          Sealed management packs can be a real pain as not all parameters of a rule or monitor are exposed for overrides. Scripts are no longer easily accessible in the console, now you have to dig into the xml of the MP to find the script. Not all MP classes are ‘public’, some are marked as ‘private’ which means that they are unavailable for targeting.

    -          Hardware requirements. The hardware to run this application is significantly more than MOM 2005.

    -          New architecture. The RMS is a single point of failover (dumb, dumb, dumb) and is very difficult to cluster. The ability to promote and demote a management server is available but wasn’t well thought through or tested enough in my opinion.

    -          Reporting. The reporting interface is not intuitive and is difficult to understand at first. It’s not uncommon to run a report that returns little to no data the first time. The report designer ‘touted as the end all and be all solution for easy report creation” is crap. I’d rather author a report in Visual Studio than bang my head against the wall with report designer not to mention the lack of report models. Yes these can be created with little effort but they should be included natively.

    -          ACS does not proactively alert you to security issues. You must still create either rules or monitors in the Ops Console to receive security related notifications.

    -          You no longer have the ability to define SMPT trap responses as an alert response.

    -          It is more difficult to report on simple server up metrics.

    -          The ability to compile an SNMP MIB is no longer supported, you must present and individual OID. This can be a real pain.

     

    Without SP1, the reliability and stability wasn’t there and the output was hard to trust, console performance was very poor and certain features (like the reset health button in health explorer) just don’t work. With SP1, the stability is much, MUCH better and console performance has also improved.

     

    I realize that I haven’t hit all of the salient points but feel free to add to the list.

     

    Overall, like with any application, it’s important to understand what you are looking to accomplish and define a list of requirements. Ops Mgr can likely deliver on the most complex requirements but for companies looking for simplistic server up/down monitoring it might provide too much functionality and although the additional functionality could be disabled, that might require more effort than you are willing to invest.

     

    For great, technical information on Operations Manager 2007 and other System Center technologies, check out our training offerings at http://www.infrontconsulting.com/events.htm.

     

     

    March 14

    TechDays Part Two {Be a hero}

    Belgian TechDays 2008 {Be a hero}

    I had the opportunity today to deliver a second session on Operations Manager 2007 Tuning and Best Practices on March 13th in Ghent, Belgium.

     

    Providing Microsoft System Center Operations Manager 2007 Tuning and Best Practices

    Level: 300

     

    This session will provide you with the ammunition to take your Operations Manager 2007 installation to the next level. If you have installed Operations Manager and are receiving unwanted alerts and feel a little overwhelmed and unsure of what to do, this session is for you. This session will provide you with the knowledge of how to perform post installation configurations, reduce noise, create custom rules and following best practices.

     

    I have posted the presentation as a pdf file to www.infrontconsulting.com/events.htm. The link is on the bottom right of the page.

     

    If you are interested in more hands-on training from knowledgeable consultants on System Center technologies, check out our course schedule.

    March 12

    Remote Operations Manager 2007

    Belgian TechDays 2008 {hero}

    I had the opportunity today to deliver a session on Remote Operations Manager 2007 today in Ghent, Belgium.

     

    Providing Remote Managed Services with Microsoft System Center Remote Operations Manager 2007

    Level: 300

     

    This session will cover the new managed services solution Microsoft is building to enable partners to create a services-based business by delivering remote IT managed services to small and midsized businesses. Supported by System Center Operations Manager 2007 (Operations Manager 2007) and System Center Essentials 2007 (Essentials 2007) technology, this combined solution creates a powerful platform with which you can provide remote software update management, monitoring and troubleshooting of customer environments over a secure internet connection. In addition, you will be able to extend the service to show your customers the value of the services you provide.

     

    I have posted the PPTX file from this presentation to www.infrontconsulting.com/mvp/ROM.pptx.

     

    If you are interested in more hands-on training from knowledgeable consultants on System Center technologies, check out www.infrontconsulting.com/events.htm.

    March 06

    The Windows Server 2003 Clustering MP for Ops Mgr 2007 is now Available!!

    The cluster MP is now available in the MP catalog!! 

    The long awaited cluster MP is now available in the MP catalog!!

     

    RTM Users

    If you are installing it on Ops Mgr RTM, remember that you will need to install the included hotfix first. (SystemCenterOperationsManager2007-RTM-KB945312-X86-X64-ENU.MSI)

     

    SP1 Users

    If you are installing it on SP1, just import it as you normally would.

     

    Note: The download page incorrectly state that the hotfix required for RTM installation needs to be obtained from PSS when in fact this is included in the downloadable MP package.

     

    If you are looking for more great information on Ops Mgr, check out our in-depth training classes at www.infrontconsulting.com/events.htm.

    March 05

    Event ID 26330 "The SDK service lost database connectivity."

    My RMS is in a "critical" state…"SDK Service - Database Connectivity"

    Event ID - 26330
    Description: The sdk service lost database connectivity. 
     
    Q. Any ideas on how to resolve this?

     

    A. The following worked for us when we encounter this. It seems pretty common… (Don’t freak out when I say put the RMS into maintenance mode. I have done this successfully.)

     

    1. Open Health Explorer for the RMS and document the monitor(s) within the Root Management Server.
    2. Disable all of the Monitors which are Critical (Red) using an Override that is targeted to only the RMS.
    3. Put the RMS server into Maintenance Mode.  (Yes really)
    4. Wait until the server is unmonitored. (empty green circle)
    5. Restart the Health Service on the RMS.
    6. Take the RMS server out of Maintenance Mode.
    7. Delete the overrides you created to re-enable the Monitors which were disabled via the overrides.
    8. Open Health Explorer and refresh the state.

     

    Interested in Ops Mgr training when you can find answers to questions like this? Check out www.infrontconsulting.com/events.htm.

    March 04

    Windows Server 2008 and RDP Connectivity - NOT what I was hoping for

    Windows Server 2008

    I want to start off by saying that I have not spent a lot of time with Windows Server 2008 but the time I have spent, I don’t consider to be as productive as I would have liked.

     

    One VERY big annoyance to me since I have started working with Windows Server 2008 is the following error message. I must admit, I have not spent a great deal of time looking into this but will turn my attention to this shortly as it’s really starting to bother me.

     

    When I RDP into my Windows Server 2008 lab server from my Vista notebook (that’s a whole other blog post), I continue to receive the following error message. The interesting note with this is that it generally occurs when I log in initially and the work around has been to RDP in from a Windows XP Pro SP2 desktop but recently this has been happening in the middle of an active session. Very bizarre and definitely worth more investigation.

      RDP session disconnect