To troubleshoot VMM issues, a debug tracing can be enabled and captured in SCVMM 2012 SP1 with the help of logman and SCVMM providers. The ETL file can easily be converted to txt file with netsh and can provide useful information on the issue being experienced. Lets go over the steps to enable debug tracing.
Eğer WMI düzgün çalışmıyorsa veya Group Policy düzgün uygulanamıyor uzun süreler bilgisayarınızın açılmasını bekliyorsanız, gpupdate/force komutunda hatalar alıyorsanız, event loglarda WMI hataları da görüyorsanız;
Hemen hemen tek çözüm WMI repository repair etmektir. Komutun nasıl kullanıldığını
winmgmt /? ile görülebilir.
Backup alınabilir, /restore ile restore edilebilir.
WMI check için kullanacağınız ilk komuttur.
Gerçekten repair için kullanılacak ilk komuttur. Eğer hata alınırsa diğer komuta geçilebilir.
Sunucu WMI resetler ilk kurulduğu günkü gibidir.
There is a great article about NIC Teaming Performance on Hyper-v.nu. Check it out:
While I was building demo/lab for network virtualization, I wanted to see what is going on behind scenes in the virtual switch. One of the features of Hyper-V is unified tracing. Windows 2012 has a new netsh parameters to be able to capture Physical, Virtual machine traffic, and traffic that traverses the Hyper-V virtual switch.
Netsh is a command line utility built in Windows OS.
One of the building blocks of multi-tenant cloud is Network Virtualization. Dynamic VLAN configuration has always been cumbersome, complex, and hard to manage. Now with Network Virtualization we can run multiple virtual networks on a physical network.
In this post, we will look at how pieces come together with the help of SCVMM 2012 SP1 and will go over creating Logical Network and VM Network setup with NVGRE.
I had mentioned about the new capabilities of SMB 3.0 in my previous blog. Lets review what multichannel buys us. With multichannel, we can increase throughput by using multiple NICs and with RSS we can use multiple cores, have fault tolerance and automatic end-to-end failover.
Windows 2012 comes with a new version of Server Message Block (SMB), 3.0, with many new features over SMB 2.0. With the new improvements made, SMB 3.0 enables server applications to use remote file servers to store their data on SMB shares, ex: Hyper-V, SQL, which allows easy provisioning, management, flexibility, and to lower cost. New features of SMB includes are:
With Windows 2012, we are now able to create vNICs in the parent partition. This plays major role in building fabrics by providing us with bandwidth management, classifications, flow control based on priority with the help of new QoS settings and functionality.
TechNet has a a common configuration guide published. I will go over the basics of one of the scenarios presented which I think will be common configuration among all implementations, 2-NIC Configuration with Teaming.
As you might be aware that NIC Teaming on pre 2012 Hyper-V Servers is not supported, which is still true per KB968703. With the enhancement came with Windows 2012, now the teaming is built in to Windows to provide more bandwidth and/or High Availability. These changes also brought great flexibility; we can mix different Vendor’s NICs in a team (well, if they meet Windows logo requirements), no vendor software needed, integrated with OS, different team options, known GUI format, support by Microsoft… Lets get to the basics…..
Trying to create VMs via SCVMM 2012 on Windows 2012 HyperV Cluster, and failing… what!!! Checking event logs I noticed System Event 5125 which did not make sense at first. The VM creation job was failing with an error but the XMLs were being created in the CSV volume. The VM properties was showing everything as expected but my new vhdx.
Log Name: System
Event ID: 5125
Task Category: Cluster Shared Volume
Cluster Shared Volume ‘Volume1′ (”) has identified one or more active filter drivers on this device stack that could interfere with CSV operations. I/O access will be redirected to the storage device over the network through another Cluster node. This may result in degraded performance. Please contact the filter driver vendor to verify interoperability with Cluster Shared Volumes.
Active filter drivers found:
After “reading” the description of the event, further investigation revealed that the ResumeKeyFilter is a Microsoft Minifilter. Full list of Minifilters are listed in the “File System Minifilter Allocated Altitudes”: http://msdn.microsoft.com/en-us/windows/hardware/gg462961.aspx
When the failover cluster mmc checked, the CSV disk was in “Redirected Access” mode. (BTW, for troubleshooting Redirected access on CSV volumes , “Ask the Core Team” has a great post at http://blogs.technet.com/b/askcore/archive/2010/12/16/troubleshooting-redirected-access-on-a-cluster-shared-volume-csv.aspx
Then with the help of fltmc.exe (minifilter driver management), I figured what the culprit was: unsupported version of McAfee. Fortunately, as one of the servers did not have McAfee, it was easy to understand what was going on:
Figure below shows 2 nodes, node1 has Mcafee installed (mdehidk). As seen, when the CSV owned by node1, it has 3 minifilters. Once the CSV volume is moved to the node 2, CSV volume has 2 minifilters attached.
And the Cluster disk becomes happy again:
By the way, McAfee will be supporting Windows 2012 in the first quarter of 2013: https://kc.mcafee.com/corporate/index?page=content&id=KB75779
Apparently, there is also a known issue for Windows 2008 R2 clusters with certain versions of McAfee (https://kc.mcafee.com/corporate/index?page=content&id=KB73596) and the solution is to install http://support.microsoft.com/default.aspx?scid=kb;EN-US;2674551.