LytchettNM

Yes you setup the server that you are going to run DA on and its wizard will add a GPO to your domain, you then add the computers we want to have remote access to this group. Gpupdate the laptop and you all done, when the remote computer has internet access outside of your LAN it will create a tunnel via IPV6 without need for the user to do anything and push any policy's you have assigned to that work station. The admin just need to bear in mind that pushing say a new package will work but take much much longer than on LAN.

Using direct access will give the network management over external computers (fully integrated into window 8 onward, windows 7 requires some extra work to the client) that are not on the direct LAN, this is done using IPV6 so some old business application that are IPV4 only will not work, but for these app's you can setup terminal services and share your older IPV4 only apps as a remote app.

Congratulations

So 2 things to remember with LAG's you MUST configure both the server and switch ports to use the same protocol I would recommend LACP without a doubt. The other thing to remember is that you won't be able to create a single transfer that exceeds the max speed of one port so with 4 x 1GB ports in a LAG the max bandwidth is still 1GB per transfer but you can now achieve 4 x transfers at a time, assuming the have the throughput from the client and servers storage. Good luck with getting the LACP config sorted.

Wow this is getting the Gary matter working which is not good this time of night! I like simple and straightforward but this depends on the network overview and hardware capabilities. I have a single subnet for our servers which terminate on the core switch as does all my client subnets basically a VLAN/subnet per building this keeps DNS clean and allows the core switch to do the networking routing and the server to just provide the services they are configured for.

Are you trying to configure the VLAN's to send down the trunk to the server? We have our core switch stack setup to route everything in the switch so I only need one subnet from the switch to the server, if you are trying to terminate all of you VLAN's on the server I think you are going to have each VLAN on the server with an IP address in the subnet for each VLAN on your server, this could be a nightmare when it comes to DNS I would say.

Just found a guide on the HP site. http://h20628.www2.hp.com/km-ext/kmcsdirect/emr_na-c03594944-1.pdf#page121 Just a quick skim read it would seem if you enable LACP on the port the switch will dynamically create the trunk port I think.

Is the switch "Aruba 2530 48G PoE+ 2SFP+ Switch" if so this is a layer 2 switch, so where are you going to do the routing for the network?

Sorry I don't know the switch so can't give you direct answers but is this switch in production or setup for testing? If not in production then maybe do a factory reset and then look at setup of the LAG again I'm sure there will be guides on the manufacturer site for basic LAG configurations.

And you only have one port assigned on the switch.

Server side looks good, but you have created a LACP link on the switch but not added it to a trunk group.

You will need to setup both ends of the bonded link to LACP otherwise to will not be able to utilise the full speed of the bonded link, it looks like you have chosen trunk and not LACP from you screen shot. When you creat the lag on the server make sure you set that bonded group to use LACP to. If you have a router / layer 3 core switch doing your routing then you should not need to tag all you VLAN into your servers, the routing device should doing that work for you.

We have the same issue and I have asked for support and have had no reply yet! But we only have a few member of staff with SIMS installed on there laptops so these teachers are aware and know that they would need to bring the laptop in and while on local network it will update. I have not find a workaround but have been thinking about setting up a remote app's server and allow member of staff that have / need SIMS at home to use the remote app. The app should run quicker to the end user as it's running locally on the server and we can update it automatically.

The best way to setup a reference/golden image is by using WDS/MDT I have just set ours up which is working well. The task sequence I've setup deploys windows 10 v1703 then installs prerequisite applications like c++ and office etc then a Powershell script removes the windows built-in apps plus checks for windows updates from our local WSUS server and pauses so I can install any programs we can't deploy it then does another pass for windows updates and then it sys-prep the image drops it into my captures folder and shuts down.

The ProCurve 2510 is a layer 2 switch so it's can do routing, you will need a router device to router between the 2 subnets.

I would try plugging a PC in on the server network (same subnet) and try to PXE from there to check if the WDS server is responding. Was the WDS server effected by ransomware? Have you got a backup of the WDS server? Why did you need to move to another host?

As I'm sure you have realised this picture show the client trying to PXE boot from 172.18.32.31 using it gateway of 172.18.35.1 so the client and server are on different subnets, and that the server has not replied. If you have a route from the server subnet to the client subnet then the issue is with the server at 172.18.32.31, if you have moved / setup a new WDS service have you enabled PXE and added a boot image?

Yes but not with cmd. Set the bios to boot from PXE first and assuming you have setup the option the remote wake these PC's and you MDT setup with the MAC address of each PC you can deploy which ever image you have within your MDT solution. Plus you can deploy other software like office within a task sequence and install all the drivers for each machine type. When your going to deploy set PXE option with WDS to always respond and during normal operation you set PXE to not respond. It will take a little time to setup MDT but it's worth the time if you have a lot of computer to service. We are rolling out the whole site this summer, it's going to be busy. BTW MDT is free.

In answer to your first and main question can you connect to subnet 10.122.40.0/24 which is within VLAN 40 then yes, but you will need to make changes to the core router which could be a switch. I would leave this alone and get the third party back in that setup your new core network as from the questioned you have asked you do not understand the way your network is setup and making changes without the basic understanding could disable your whole network. The simplest thing to do would be to get VLAN 40 presented to a switch port on the switch that feeds your support location and have a dedicated computer that you make these changes from, this is a much simpler things to achieve with limited networking knowledge and understanding.

Have you tried this method? Try performing a factory reset using the instructions below: NOTE: Factory default button will be held down for a total of 90 seconds. While the AP is turned on, push and hold the reset button for 30 seconds. Turn off the AP, continue to push and hold the reset button, do this for another 30 seconds.

I like that idea. I think I might take that one on myself thank you.

I would expect BYOD to be a large subnet so that's not an issue, 30 day lease that sounds a little long to me we run a 6 day lease and auto scavenge every 4 days with 4 day update. Keeping DHCP and DNS heathy make for a stable network especially if the network grows.

Having looked at you whiteboard [emoji106] it very wasteful of IP ranges if I've understood it correctly, you have 10.40.x.x showing BSFC but after 10.40.6.x or then jumps to 10.41.x.x assuming that this is not the first of your other sites I would not do that.

We also standardise on things like all gateways they are all x.x.x.254 and most networks are a /24 network aiming to use about 100/120 ip per scope and devices that need a fixed IP they start from x.x.x.201 within the subnet they are within, printers, switch management, CCTV etc are in their own campus wide VLAN/subnet and things like WiFi we then use a /22 network.

So looking at just the IP side of things a simple but flexible IP structure has got to be a baseline for today's MAT networks imo. Something like this:- Site 1 10.0.x.x Site 2 10.1.x.x Site 3 10.2.x.x Etc. We use VLAN/subnet to keep broadcast under control and again to keep things simple we use the third octet to reflex the VLAN id which can make fault finding easier and standardise network configuration across all sites. So if you have a server solution at each site you could use the same VLAN/subnet at each site for servers and another for network management etc, this will help from a network support prospective and for that matter setting up ACL's which you can use to block student and staff networks from having access to building or switch management etc. I find the key to an efficient / stable network is to keep broadcast under control, use spanning tree, standardise the way networks are structured and restrict who has access to make changes to the core / edge networks.