kmount

Corsair K70 with cherry reds.

Just looked on two different customer systems (running a virtual veeam server) and both have it there. (one is Veeam 9u1 and the other Veeam 9u2)

To be honest you could just uninstall update manager, nuke the db and install it again from scratch and it'll go away and download the necessary and scan the hosts. An upgrade of SQL to 2008R2 to get the 10GB limit would be recommended in line with making sure the VMware side is compliant (i.e. patch levels of vCenter with SQL versions) with the interoperability matrix.

Hey Ant, Thanks for this. It looks like it has all sorts of knobs and screws to get a real good fit, I'll be on the look out for one to have a go in tomorrow when I'm on the great chair hunt! Cheers, Kim

Hi there, I'm looking for a new office chair. What do you guys use? It needs to have very good back and head/neck support when sat in the normal (non reclined) position. I will spend easily 10-12 hours in it a day. I've been very disappointed with the quality of the humanscale freedom chair that I'd hoped would have been something good but it seems to insist you recline when you want any form of support. Thoughts welcomed! Thanks! Kim

Hi, Recommend you look at - https://msdn.microsoft.com/en-us/library/azure/dn151324.aspx In short you probably want 4 if you're doing it properly (ie. proxies in DMZ, main federation servers internally) and have 1000-15000 users but you may find you could do it co-existing on domain controllers for under 1000 but I'd be wary of this approach if you ever wanted to change your DCs you need to move ADFS etc. All explained in the link above.

Hi, This is likely because of nested virtualization (i.e. a VM inside a VM) I've done this the other way around, running Hyper-V within a VMware VM. https://msdn.microsoft.com/en-us/virtualization/hyperv_on_windows/user_guide/nesting might help.

Run pathping 8.8.8.8 and see where the loss is.

We set this up for a customer, works well. We hit a bug - https://blogs.msdn.microsoft.com/virtual_pc_guy/2014/06/11/allowing-non-administrators-to-control-hyper-vupdated/ even though our Windows 10 wasn't an upgrade we hit this!

Much better CPUs!!

I think you've landed in the middle of a MBR/GPT battle with the 2TB limits. Can we see a screenshot of disk management please. Is this a single 3TB RAID volume and you've installed windows onto that or is windows on a separate boot volume?

For 40-50 you should be fine virtualising the whole lot, just watch the clock speeds don't devalue your solution. Not a lot of point buying tin for that small a number, just works out expensive. Go with localzuk's model, it makes sense and similar size environment.

As localzuk said it's all down to what you're doing with them. As someone who does this kind of stuff for a living similar to your consultant it may be them erring on the side of caution if your hosts are already busy doing other things it takes away potential resources from your new RDS environment. I'd be a little wary of your clock speed only being 1.6GHz it's rather low - won't stop you doing it but is something to keep tabs on if your users are likely to be hitting the clock hard. (not so likely as you said Office, Internet usage on the most part) Typically we find when you hit sort of 200/250 concurrent connections it starts to make sense to consider a small number of physical servers vs a truck load of virtual ones to get the same kind of density. We support both kinds though; it's alright until you end up managing a large number of 2/4vCPU RDS hosts when you could run a small number of higher user density physical core ones instead. (yes, the user impact of a host failure is larger but the management time is lower and you're always at least n+1 right? and if you virtualise more than 1 client server on the same hypervisor host as others the argument is similar anyway) Devil in the detail though, how many concurrent users are you expecting and if you're expecting youtube/etc using RemoteFX I think you may start to hurt that low clock speed (web browsers and multimedia content hurt) - make sure you protect your other infrastructure running on there from the RDS servers too! As localzuk said, you may want to consider separate VMs for the management environment - if it's only small numbers you may be able to write the redundancy risk off as a noted issue otherwise you need a pair of gateways, connection brokers, sql servers (with witness) and licensing services. For most small environments though you can get away with 1 of each (and no SQL) and in really small ones some people will co-exist some of the roles together. Having it virtual helps you not necessarily need to do that though.

Need more information on what exactly it is doing, it should tell you in the summary window - I presume it's manipulating the restore points and it's taking time. How big is the resulting vrb (your latest backup is the vbk, I'm interested in the most recent vrb AFTER the job is finished) What's the backup storage target? Do you backup the vCenter via the vCenter or via a direct host connection? How many restore points? What compression/etc settings or default? Does a brand new job on the same VM complete fine suggesting the issue is the backup chain/length etc

Treat with pinch of salt but I got some pricing back in January for this for an existing customer of ours looking at doing similar. I was working with an Edu customer on sizing out 25TB and cost on RRP was ~£1530 pcm. (no addon costs for bandwidth, flat rate on storage) (hosted in a UK DC, peered with JANET etc) Our customer decided to utilise another physical location some miles away instead.

Link here to compare what each edition includes functionality wise - Compare vSphere and vSphere with Operations Management Editions | VMware United Kingdom

Hi there, Yes, you can expand this using an RD Connection Broker which will help introduce load balancing between your servers. Are you truly providing a virtual desktop (i.e. using Hyper-V) or are you using RDSH (session host, single instance of windows shared by all users). If the latter then yes, just create another server with the same applications etc and create a load balanced farm (with RD Connection Broker) and you should be onto a winner.

OK that rules out DNS shenanigans. What actually happens when you try internally, what do you see?

Hi, Ping it from home note down the IP that it returns. Ping it from school, compare if the IP is the same or not. (it may be that the external provider changed the IP of the site you're trying to access and because your colleague added a forward zone for it it won't necessarily know of the update)

With 9 hosts you cannot use the 'Essentials' line unfortunately. (max of 6 CPUs) You'll probably find Standard is enough for you unless you want some of the more whizzy restore options in Enterprise. I cannot imagine you needing Ent Plus unless you're looking to integrate tightly with storage, or WAN accelerators etc. Good comparison of editions at - Editions Comparison – Veeam Backup & Replication

What's the switch saying about life on those ports?

Hi, If you provided them with the CSR then no as they won't have your private key. If you used a tool of theirs (an automatic system on their site) to generate the actual CSR+Cert/etc then they may provide you with a pfx file which can have the private key extracted from that. I suspect you would have provided the CSR from a machine that you generated it on (this is the machine that will have your private key on).

Check out 1) https://technet.microsoft.com/en-us/library/cc753380.aspx for your setting of a max limit per server 2) https://www.microsoft.com/en-us/download/confirmation.aspx?id=17190 for sizing and such, they have a single server for 450 users mentioned in there. Typically you exhaust your physical resources before you exhaust the limits of RDS. (note above based on 2008R2 but good place to start)

Yep, the dynamics of USB have improved over the years with RDSH but at the end of the day it's still a 'shared' OS unlike in RDVH. RemoteFX will make the difference night and day. Windows endpoints will have the best capabilities, coming from a Linux guy. So using something like thinpc or cut down kiosk'd windows is a good way to go.

USB is normally the big one these days, RDSH support is much patchier than RDVH. RDSH is more cost effective to deploy so where possible starting there and seeing if it meets your needs makes the most sense. You can blend them together though if you found a specific need for some RDVH (VDI). 2012R2 + RemoteFX is quite a compelling set up, we ran a POC with a customer recently by adding Nvidia GRID cards to their RDVH hosts and performance was spectacular. Reducing your C2D/2GB jobs as repurposed thin clients is a good angle too to save you beans.