@Kevin - Thank you a lot I do really appreciate it!! I hope there will be a new version available soon.
RHC
26 Jun 2012 1:26 PM
28 Jun 2012 5:11 PM
Great write up Kevin. I'm looking for the previous version (6.0.6958.0) of the OS management pack. I installed this in my lab and need to revert back, but someone cleaned up the repository where we kept previous versions. Any ideas on where to locate previous versions?
28 Jun 2012 6:18 PM
Ugh - getting access to previous versions of MP's can be very difficult, and Microsoft no longer makes them available. You can probably open a support case and request them if there is a bug that is a showstopper for you. Alternatively, you can shoot me an email request at kevin.holman@microsoft.com and I'm happy to shoot you what you are looking for if I have it.
RHC
5 Jul 2012 3:06 PM
Hi Kevin
Any news/updates on the cluster resoruce and cluster disk monitoring issue?
I have a microsoft case running on the same problem, but I will like to know if you have any updates?
5 Jul 2012 3:30 PM
@RHC - no updates from me. At this point it simply comes down to customers providing direct feedback via cases, if this behavior is a concern for them. I am not in the product group, direct customer feedback via a support case is the best way to effect changes.
Sjoerd
10 Jul 2012 10:32 AM
Please put pack in the alert description
The values that exceeded the threshold are $Data/Context/Property[@Name='PctFree']$% free space and $Data/Context/Property[@Name='MBFree']$ free Mbytes
its sucks to look at something that says . there is not enough space, but doesnt tell you how much, and i wanna keep using sealed managmentpacks instead of tweaking my self.
11 Jul 2012 12:15 AM
@Sjoerd - did you read the article? I explain the new functionality in the new monitors which gives you this back, or the ability to use my addendum MP.
RHC
19 Jul 2012 3:29 PM
Hi All
I have had a Microsoft case open on the cluster disk monitoring problem (the problem where it only monitors the cluster disks of the resources that is located on the same node as the quorum resource)
As Kevin already mention Microsoft is aware of the problem, but they have told me that it will be fixed in the next release of the Base OS MP – This MP will be released together with Windows 8 and Windows server 2012.
::: Here is the information from Microsoft :::
INFORMATION / RESOLUTION
The issue is that the new cluster disk discovery is placing all disks as hosted by the quorum resource group (or possibly the first resource group discovered). Therefore, we will only monitor the cluster disks when they are on the same node as the quorum resource group, which should not be the case as they are independent, especially in multi-node clusters or active-active configurations.
This issue has been reported to the product group and they confirmed that they will fix this on the Windows 8/Server 2012 timeframe, when they will release a new version of the BaseOS MP to support Windows Server 2012. They will do their best to release the MP on the release date of Server 2012, if not they’ll do it sometime after the launch.
Mark Verbaas
23 Jul 2012 6:18 PM
Hi Kevin and others,
why did Microsoft decide to split the free space monitors in two monitors? The 'old' monitor covered the larger and smaller disks, because both clauses must be met (Free MB and %). Now there is a need to create a whole bunch of overrides. Is there anything we can do to enable the 'old fashioned way' of monitoring our disks?
Regards,
Mark
23 Jul 2012 6:29 PM
@Mark Verbaas
The "old fashioned way" is enabled by default (for logical disks). These new monitors are not enabled by default. The reason this was done is twofold:
1. To enable consistency for how cluster disks, and CSV's were monitored. 2. Many customers never read the documentation, and did not understand that BOTH thresholds needed to be met, and would not get alerts when they expected. By splitting this into TWO monitors, they easily get access to Percent, and/or MB Free, which is much simpler to understand.
There is no easy answer which makes everyone happy. The expected future is to utilize the new monitors, which is two independent monitors with an aggregate rollup monitor. This is the most flexible design offering simplicity, and flexibility to allow people to use a single unit monitor, both, or optionaly enable to aggregate monitor, all while allowing for ticketing systems to work with creating warning alerts, then receive critical alerts from the aggregate.
The only issue i see, is if you really liked the previous design, there is no option to utilize it for Cluster Disks or CSV's anymore.
Mark Verbaas
24 Jul 2012 9:14 AM
@Kevin,
Thank you for your reply. We really learned to appreciate the single monitor approach. We'll figure out a good set of thresholds for the new approach.
Thanks again!
RHC
24 Jul 2012 10:03 AM
Hi Mark and Kevin
@Mark I totally agree with you, as I wrote earlier I really like the “old fashioned way”. I do also understand why Microsoft decided to take the other approach with 2 monitors, and I like the why they decided to implement it for logical disk (it’s up to the customer to decide which one of them they like) but I don’t get why they didn’t implement it the same way for the rest of the disk types.
I do really hope that this will be changed in the next release, therefore I have send a “request for change” to Microsoft on that issue - @Mark and others that wants the same change please send the information to Microsoft so we can show them that there are customers that like them to implement the 2 possibilities for all disk types.
Mark Verbaas
24 Jul 2012 10:13 AM
@RHC,
Did you create a feedback item on connect. If so, please post the link here so we can vote on it. Otherwise, we migth consider creating a feedback item.
RHC
26 Jul 2012 4:19 PM
@Mark - No I didn’t do that, honestly I don’t know how to do that, can I ask you to do it and then post the link?
Ramesh
28 Aug 2012 1:41 PM
Kevin - Another thing noticed in this new Base OS MP is that the threshold for "Available Memory Threshold (MBytes)" parameter is 100 on "Available Megabytes of Memory" performance monitor whereas threshold for the same parameter in the old Base OS MP version 6.0.6794.0 is 2.5
What is the difference?