System Center Operations Manager 2007: Tips

PerfDataSource - 10102 / 1103 - Missing Performance Counters

Thu, 03 Apr 2008 18:07:32 GMT

It can get frustrating when working with performance views and some servers have all the expected values while others don't even though they are the same OS etc. When an agent has rules to collect various performance data, but the performance object and/or counter is not available it generates an Error in the server's Operations Manager Log. Event ID 10102 and it is always followed by a Warning Event ID 1103. It typically logs:

Event ID 10102 starts off with "In PerfDataSource, could not resolve counter Process, Handle Count, HealthService*. Module will be unloaded blah, blah blah" while Event ID 1103 starts off with "Summary: 1 rule(s)/monitor(s) failed and got unloaded, 1 of them reached the failure limit that prevents automatic reload. Management group, blah, blah"

All this means is the server's operating system has a problem with its perfmon counters, they are either missing or corrupted. Go check it out, I'll wait.......see I told you. :-) It is not an OpsMgr issue. What I do in almost every deployment is create my own Event view to display only 10102 and 1103. Run a report and send it to your friends on the server team.

Exchange Management Pack Configuration Wizard

Sat, 29 Mar 2008 20:50:21 GMT

Running the configuration wizard in small Exchange environments is very simple. For example, a single site, single domain environment will not present much of a challenge, but what about multi-site, multi-domain Exchange organizations? What is often overlooked is the role Active Directory replication plays in the successful configuration of the Exchange MP. I recently spent about two days yanking my hair out because absolutely nothing was working with respect to configuring Mailbox Availability / Mail Flow. What I soon learned was my Mailbox Access Account (which was created about 10 minutes before running the configuration wizard) had not replicated to all the local domain controllers for the various Exchange sites. It turns out the client has some funky replication settings so it took almost 3 days (an entirely different set of headaches...). Once replication succeeded, all was well. Here are some tips on making the configuration of the Exchange Management pack simple, even in complex environments:

1) Mailbox Access Account - Create it in advance and verify replication has succeeded. Spot check the DCs the Exchange servers (back-end) normally query.

2) Do not use Default configuration type, choose Custom so you can configure one piece at a time. Once verified it is working correctly, simply re-run the wizard. I normally start with Mailbox Availability (Per Store Monitoring) and give it 24 hours before worrying about errors. I know this sounds insane, but it is well worth it, especially with global networks with DCs and Exchange all over the place. I then move to Mail Flow. Discuss with Exchange Admins' regarding Per Store vs. Per Server. Simply turning everything up is sometimes one colossal headache. A practice I have been adopting is looking at the stores with the slowest MAPI Logon Latency and then designing an effective Mail Flow. Mail flow should mimic the real time flow of mail too.

3) Front-End Monitoring. Rather straight forward.

Of course Message Tracking is enabled too so all Exchange Reports function correctly.

Exchange White Space

Fri, 21 Mar 2008 19:42:11 GMT

http://ianblythmanagement.wordpress.com/2008/03/21/exchange-white-space/ | Comments

This is something that Exchange admins ask for. The trouble is that the number is buried in the event description.

OpsMgr has an event rule already that collects these events. The rule is called “Free space available in databases after online defragmentation.” and the event ID is 1221 with source “MSExchangeIS Public Store” from the application event log. You get a sentence in the event description like

“The database “XXXMSX02 SG1\XXXMSX02 SG1PF1″ has 63 megabytes of free space after online defragmentation has terminated.”

What I did was create an event view for that ID and the admin can just quickly go down the list for that day and look at the numbers in the description. People that know scripting may be able to do something funky to collect the information and alert on it depending on the size of the number. But I will leave that for someone else to do. The Exchange admin was happy with the event view as it collated all the information in one place and made it easy to see the details.

It would be better if the Exchange dumped that into a performance counter or WMI so that it can be look at and alerts created.

Object Discovery Frequencies

Fri, 21 Mar 2008 15:35:05 GMT

http://RoryMcCaw.spaces.live.com/Blog/cns!2C4CD58BB2826E21!3064.entry | Comments

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.

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.

Best Practices for Disabling a Rule

Thu, 20 Mar 2008 13:37:20 GMT

http://feeds.feedburner.com/~r/systemcenterforum/~3/245102591/ | Comments

With the temporary workaround for the ‘Performance Module’ issue requiring disabling of a rule, it’s got me thinking about best practices for storing your overrides.

Issue: Everyone knows that you can right click a rule in Operations Manager 0r Essentials and create a rule disable override. However, when you select the Disable option, you will not be prompted for the management pack, meaning the disable override will be stored in the Default Management Pack. Not ideal.

Workaround: Instead, 1) Select Override the Rule, 2) Select the target object(s) and the MP where you’d like to store the override and finally 3) Override the Enabled property and set to False.

Same effect without touching the Default Management Pack with your overrides. While you’re at it, back up your unsealed MPs. Just as in MOM 2005, backing up your customizations prevents the need to perform a database restore to retrieve customizations deleted or otherwise mucked up by someone else.

Update: Checking Operations Manager 2007 Agent and Server Versions via Powershell

Thu, 20 Mar 2008 13:35:52 GMT

http://feeds.feedburner.com/~r/systemcenterforum/~3/187869372/ | Comments

3/2/2008 - Updated versions throughout for SP1 (RTM) released in Feb ’08

Although you generally check Pending Actions in the Console UI, the built-in cmdlets provide the means to easily retrieve agent and management server versions.

Returns Name, Version of all agent-managed machines
get-agent | ft DisplayName, Version

Returns which agents are not yet at SP1(RTM)
get-agent | where-object {$_.Version -notmatch ‘6.0.6278.0′ } | ft DisplayName, Version

Returns which agents are at SP1
get-agent | where-object {$_.Version -match ‘6.0.6278.0′ } | ft DisplayName

And of course you can do the same with your Management Servers
get-managementserver | ft Name, Version

Finally, you can also use Powershell for bulk approval of Pending Actions to update your agents with the Get-AgentPendingAction and Approve-AgentPendingAction cmdlets in a one-liner.

Get-AgentPendingAction | where {$_.AgentName -like “yourserver%”} |Approve-AgentPendingAction

Versions:
RTM = 6.0.5000.0
SP1 (RC) = 6.0.6246.0

SP1 (RTM) = 6.0.6278.0

SystemCenterForum has partnered with Infront Consulting to bring you the Operations Manager Bootcamp! Register in December for training in 2008 at SystemCenterForum and receive a free copy of Operations Manager 2007 Unleashed!

Event Log Monitoring and Backup MP

Wed, 19 Mar 2008 13:34:15 GMT

http://feeds.feedburner.com/~r/systemcenterforum/~3/245005546/

From Andrzej "IT Thoughts", Management Pack for monitoring Event Log size and performing Event Log backups if needed.

Powershell Tip: Operations Manager 2007 Top Alerts Report - Part 1

Tue, 18 Mar 2008 12:58:23 GMT

http://feeds.feedburner.com/~r/techlog/~3/253454235/powershell_tip_operations_mana | Comments

Pete from SystemCenterForum: My favorite characteristic of Command Shell is the availability of the huge array of rich formatting, filtering and sorting options. Oh, and the fact that I can do almost anything in one line. Put these two things together, and we can do some pretty amazing stuff. Watch as we take a MOM 2005 function with it’s own Solution Accelerator (the Alert Tuning SA), complete MOM Reports - and deliver a functional substitute in a single line of Powershell code.

Let’s have a look at the solution and deconstruct what we’re doing. In addition to the obvious get-alert cmdlet, we’re using 4 additional cmdlets together here, each of which has a purpose in filtering, sorting and formatting the output.(continue at source)

Part 5: Distributed Applications and Xian Network Manager IO

Mon, 10 Mar 2008 18:58:14 GMT

Please check out http://www.systemcenterforum.org/part-5-distributed-applications-and-xian-network-manager-io/ for additional information. XIAN is one of my favorite products and has proven invaluable in true end to end monitoring solutions.

Update: Checking Operations Manager 2007 Agent and Server Versions via Powershell

Mon, 03 Mar 2008 03:52:37 GMT

Courtesy of: Pete