The VMM team wants you…to tell us what you want

The VMM team wants you…to tell us what you want

  • Comments 21
  • Likes

imageOn the product team we get excited to know people are using the software and services that we build every day. We joined this team because we want to make your life easier. We know a lot about our product and through this blog we hope to teach you more about it. What we don’t always have a lot of insight into are the challenges and problems that motivate you to deploy and use our software or a competitor’s solution, so having more insight into the challenges and problems you face will help us better focus and prioritize on the solutions that we need to build.

If you are interested in influencing the future of Windows Server and System Center, join the Microsoft IT Pro User Panel.   Panelists have the opportunity to provide vision and feedback to the Windows Server and System Center product team through surveys, interviews, focus groups, usability sessions and early design concept reviews. Areas of expertise we are looking for are Infrastructure-as-a-Service (IaaS), virtualization, Windows server administration, information security, automation and mobile and computer device management. You do not have to use Microsoft products to qualify for our panel. To get started, please take our user panel background survey here: http://aka.ms/ITProUserPanelSurvey.

Thanks,

The VMM Team

Your comment has been posted.   Close
Thank you, your comment requires moderation so it may take a while to appear.   Close
Leave a Comment
  • To my experience you should focus on stability and reliability on this product.

    Please have a look here:
    http://www.atumvirt.com/index.php/2013/09/it-cant-just-be-me-a-rant-on-hyper-v-and-scvmm-2012s-failings/

  • Making it so that we only have to use one tool to manage it all would be nice. Today we have to use SCVMM for some things, HyperV manager for others, and Failover cluster manager for others. And some things we can only do with powershell. And it should go without saying that all of that should actually work as opposed to allowing you to click "submit" or "ok" and then fail anyway.

  • Not having management of Hyper-V replica within VMM has meant that we have had to implement a 3rd party replication system for the last three Hyper-V 2012 customer projects I've worked on. I think the Azure Recovery Manager offering is too expensive to be taken seriously at volume, nor is there an Azure presence in my country (yet!) - Australia.

    My other big gripe is how slow VMM is with initiating jobs on clusters. At the last environment I built (2012 R2, both VMM and Hyper-V), a 10-node cluster takes about 30-40 minutes to refresh. Putting a host into maintenance mode (running 120~ VMs) takes upwards of an hour. But draining it via FCM takes about 10-15 minutes. I can only imagine things will slow down as the customer scales out with more nodes.

    The impact of the speed issues is that the customer ends up using both VMM and FCM for basic operations that they really shouldn't have to leave the VMM console for. And this creates a poor perception of the product in the trenches.

    Yet there are some things I really love about VMM + Hyper-V. The Powershell automation is superb - "View script" is absolutely fantastic and has enabled me to rapidly come up to speed with automating things much faster than if I had to read Technet API doco.

  • Things I'd like to see:
    Hyper-V replica
    VM migration that works
    UI that doesn't crash
    A place to see relationship between VM networks and logical networks
    A built-in software load balancer, and a gateway, or at least a built-in template
    Update management for VMs
    Support for cluster library server
    Better networking management for clusters, I don't want to update node by node


    Things I don't want to see ever again:
    Unsupported cluster configuration
    Jobs that fail because another job has a lock on some object. Never says what object, what job
    Agent installation failed prematurely. Good luck troubleshooting that .
    Duplicate VMs in a cluster
    Missing VMs in a cluster
    A "repair-VM" action that doesn't do anything
    Jobs stuck at 99% (new VM)

  • A better template setup would be nice. The current setup may work well for service providers but I don't think it works well in smaller environments. I would like the security sets to work through all management interfaces the same (VMM,Appcontroller,Azure pack).

    There really needs to be a better connection between what the cluster is doing and what vmm thinks is happening. There are too many times when vmm is not in sync. (2012 R2 is much better with this)

    Nothing is more irritating that getting a error message that tells you nothing but a message saying contact your administrator. Just give an error that is descriptive enough to fix the problem quick.

    It would be nice to see the ip info on the virtual machine stats window when in VMs and Services. Maybe the entire network chain from ip->nic->virtual machine network->pool->logical network->logical switch.

  • Things i am missing:

    - Hyper-V Replica support
    - Ability to move multiple VMs in a cluster at once - i can do that in Failover Cluster Manager but not in VMM
    - Consistent Naming, "SCVMM xxx Resources" in Failover Cluster Manager is just creepy

  • These are all helpful, thanks. Please consider signing up for the user panel. The link is in the blog post.

  • Funny... I did the survey and they never ask for any contact information. But at the end of it, it tells that MS will contact me if I'm chosen..
    How interesting !

  • SCVMM 2012 R2 - I just dont trust it. I dont trust it to do anything i ask it to do given the more or less 50/50 change of success. Its a terrible product. I agree with everything Marco said in his rant, and with various other posts that are HEAVILY critical of this product. Its more of a hindrance than a help, to be sure.

    I wish that Microsoft would put some 'A' team developers on this product for a while,....

  • I'm going to have to agree with most everyone else...I tire of having to leave VMM to go to Hyper-v manager or Fail over cluster manager or log on to the physical box itself to see/do things.

    Why cant I add virtual network adapters to the cluster and have that propagate down to the nodes and then why do I have to run my vb script to change the bind order and other network settings on all of the cluster nodes.

    I can hot add disks to a vm and grow the size but not memory or cpu without a shut down...myself and the users find this annoying.

    I have had to tell to other admins "not to believe what you see in VMM" to go to FOC manager before they call me with an "everything is down". "host not responding"? It seems to be responding to everything else why not VMM

    The library sux...I have a nice folder structure organized in a manner which suits me fine, but if I go to add something out of the library I get a reorganized jumble of bullshit to choose from...why cant I right-click on objects or drag and drop to/from the library to do things. There are too many library rants for me to post them all here.

    I will certainly agree with everyone else that these days VMM is painful slow and unresponsive. VMM 2007 was snappy and responsive since then it is not.

    What happened to multi select? If I try to multi select it just clears itself after a couple seconds...Ugh. The list can go on.

    We all seem to have the same rants about VMM...there are a lot of great features in VMM to me they just seem kinda half thought out...So, unless Microsoft itself is using VMware for it virtualization platform there must be VMM and server admins at Microsoft that have the same pain points as the rest of us in real world land...Maybe the VMM developers could walk down the hall and have a conversation with an actual admin...not his manager that hasn't done an admin job in 20yrs or maybe even not at all...and work on completing the vision.

    -Enjoy

  • - Refresh of the NVGRE policies still flaky.
    - I dont want to mess in the database to remove a dead host yes please build a real -force
    - I cannot deploy a Logical switch using abolsute bandwidth setting... need to do it in powershell.
    - VMs Usage metric collected by SCVMM seam not always accurate and take long to refresh.

  • Reporting without the need for SCOM. The database is there, give us the data in VMM.

  • To add to the list of already suggested features: Enhanced Session Mode support.

  • P2v back in, Replica support, multiple selection for migrating, online upgrading of integrated Components then just restart instead of offline for 10 minutes or so.

  • I have to agree with what has been said before..especially this :

    Things I don't want to see ever again:
    Unsupported cluster configuration
    Jobs that fail because another job has a lock on some object. Never says what object, what job
    Agent installation failed prematurely. Good luck troubleshooting that .
    Duplicate VMs in a cluster
    Missing VMs in a cluster
    A "repair-VM" action that doesn't do anything
    Jobs stuck at 99% (new VM)

    also:

    make sure you won't ever a new feature that only in hyper-v manager. QOS is a good example.
    make sure new features are supporter across the board, no gen 2 in service template..
    VMM has to be snappier..and more efficient
    also, WMI has to be much more robust and reliable that what it is right now..
    Dashboard are currently lacking a lot.. please add some trending in there.. I want to have everything in VMM, scom is not an option..
    why not put some undersized - oversized vm reporting capability in there..
    vcenter operations manager is probably a good place to start looking