KK20

bloody salesmen. At least I know one company that is on my do not accept calls from now.

this is what im trying to find out. I have never heard of nor can find any information or caveats to exhale costs. I am using the native app on synology and qnap to "backup 365", this simply connects to our tenant, I choose which sites or roots I wish to backup and away it goes. Cheaper than paying a cloud to cloud provider since I already have the NAS. That being said the salesman was talking about bills that appeared to various clients citing exhale costs! Just thought I would throw it out there to see if anyone else had heard of this before.

Had a salesman trying to sell me things, usual cold calls. However, he kept going on about education 365 "exhale" costs for backups and was warning that we were going to get a substantial bill (quoting this happened to Phoenix and Veeam). We use sharepoint and onedrive for our storage, we arent a huge consumer and have around 800Gb of total storage. we repurposed our old QNAP and synology NAS boxes to perform a full then incremental backup of our 365 - this is using native apps on qnap and synology. These are in place for disaster recovery only, I am not expecting 100% uptime backup and our risk assessment is happy enough with the nightly restore AND the time taken to do so. We have never been throttled for our access, I use a dedicated 365 account for each NAS box backup. I have zero idea on "exhale" costs and have no idea if this guy was peddling snake oil or not. Any ideas out there?

I had the same email on 23rd also.

I just went for it we havea couple of admins to reset my account details if it all went wrong..... I did have difficulty at first as it seemed to want me to put at least one URL with the isams cloud pattern for the enterprise app. I simply added a fake one, added my real one, set my real one to default, removed the fake one. This seemed to let me save. You do need to update the authentication mechanism on each isams account though but other than that it works just fine.

we migrated from E1 to isams some years ago. It was hell on earth. Raw pure hell. In that time Isams support has gone downhill steadily. I need to complain about every ticket to get it sorted now, the first line support just send a quick boilerplate answer - usually along the lines "internet error get more bandwidth", then low and behold an update a week later fixes it. For a huge chunk of time it wouldn't work on 1024 screen resolutions. A poor resolution perhaps? Yes but not for someone with visual issues on a large monitor - their preferred way of working is to set a fixed resolution rather than need to hotkey zooms (after working with the person it becomes quite clear that it can be a real pain getting some applications to "zoom in" or work with a magnifying glass, it is much easier to lock the screen and text size), and isams DID support that resolution both before and after they upgraded the UI (they forgot to put scroll bars on the tabs so you couldn't cycle tabs not on the screen). They blamed the user on using an old screen resolution - that went down well (the fix on our side was a huge TV on 720p which worked at the time but looked very silly on the desk!) The best ticket ive had from isams has been "we dont support office 365 for email". Pardon? So you dont support one of the largest email servers in the world? Luckily I needed to jump through such hoops with an older UPS so simply used the same relay and added another firewall exception for isams. Zero help at all from isams other than "we can sell you a module that should work". I would say that number 1 on everyone's list should be "quality of support", an MIS will live and die on both fixing a problem or asking a question on "how do I do xxx". More and more, isams answer is becoming "pay for a trainer" when asking a question. They have almost zero documentation now, I hoarded a lot from the early days when the support pages DID have documentation. Luckily I download web pages when I find a good solution to something. For this reason alone I advise people to think twice about isams now so it isnt surprising that they are dropping down the lists when business managers are being told "to fix X you need to spend on a module", Bursars notice when the annual bill goes up a lot

We have onsite isams, it is currently using classic AD logon for our users (using isams 2FA). We run hybrid AD and leverage 365 SSO for just about everything else. I was looking at https://docs.microsoft.com/en-us/azure/active-directory/saas-apps/isams-tutorial and was wondering if anyone had tried this for onsite installations? I might "just have a go" later but I couldnt find any information on people who had done this locally. I was just going to swap "https://.isams.cloud/" with our regular isams URL

holy necro batman. We use SetUserFTA as microsoft just wont let go of PDFs for us, no matter what GPO I set, how I configure, after updates Microsoft Edge will randomly seize control of PDFs. I have a config file with and I simply call setuserfta with this config file. Seems to work for us.

The difference with W10 and W11 on old hardware is the threat of withholding updates and patches.

This is where you will have fun depending on your CPU (im assuming you are using the onboard VGA that is baked into the CPU). You will need a USB docking/port replicator, these have built in VGA, we use them as "cheap docking stations" for our laptops like these https://uk.dynabook.com/discontinued-products/pa3927e-1prp/ If you want to "extend" then you wont have a problem. If you want to "clone" then you will as you are down to intel drivers supporting clone - I think 5th gen and above support clone with external display. The workaround is to disable the intel adapter and let windows install the basic adapter.

My strategy is bury my head in the sand for a few years and hope I get a budget increase. We still havea lot of latitude gen4 i5's with replacement batteries. I have a handful of desktops that are circa 2013 that are connected to various single use applications (the laser cutter, milling machines, boiler management terminals etc). I would estimate at least a half of my machines wont run 11 effectively. The core desktops are all 7020 i5 with 8gb and SSD, they run 10 brilliantly but wont support 11. So im screwed really and I only retire in 12 years so I will have to fix things :-) I mean, I got core2duo's working just fine on W10 1604 and those started life with XP on them...

enquire if you are eligible for "K-12 SCHOOL SITE" it is about £23 per year per device for shared device model. Minimum seat of 25. You get the full CC, you will need to set up azure AD sync and login for each user.

We had an issue with infocus projectors doing a similar thing (around a dozen classrooms, different laptops). Our viewsonic projectors did not exhibit the same effect. We ran a new 10m VGA cable and it worked fine. I suspect the original cable was either too long, in a duct with power cables etc or something else. I was lucky enough to be able to cable tie and run a new cable though.

I personally find powershell to be a bit heavy for startupscripts, I tend to use VBS which are a bit lighter and seem to execute quicker. If you want to add printers using VBS then Set WshNetwork = CreateObject("WScript.Network") WshNetwork.AddWindowsPrinterConnection "\\\" WshNetwork.SetDefaultPrinter "\\\" set wshnetwork=nothing works quite well. If you wish to delete printers or clear things out then: strComputer = "." Set objWMIService = GetObject("winmgmts:\\" & strComputer & "\root\cimv2") Set colInstalledPrinters = objWMIService.ExecQuery ("Select * from Win32_Printer Where Network = TRUE") For Each objPrinter in colInstalledPrinters objPrinter.Delete_ Next obviously this will not bypass the admin requirements as a "user" but you can execute these under whatever security context you choose. I simply find VBS to be quicker.

I wasnt happy with disabling the restriction via the reg edit (our point and print restriction had already been set in the past). So we are stuck with kyocera type 4 universal drivers. The problem with those is the lack of device control for things such as forced monochrone etc. That is where we are at for the moment. I did try installing the type 3 driver locally first (via startup script), I can verify the type 3 driver is present in the print management->drivers page, however, if another printer is added (with the same driver) then the prompt appears and needs admin credentials. so it seems that even with a type 3 installed I still get the prompt.

This is how we phased the migration. 1) informed users and ran a few session on what we were doing and WHY. 2) used "cloud drive mapper" to map H: to everyones onedrive (whether they were in the trial OU or not) 3) picked a trial OU of users. Set their "my documents" to read-only. 4) users could use their documents for reading but had to save changes to their H: drive. 5) once all OU were part of the "read only local" restriction, give people 6 months before their documents were switched off. 6) switch off the "my documents" redirect. Set "my documents" to be onedrive. Leave H: because it is ingrained in their usage (but in reality can use both) 7) perform 2x hard offline archive backup of the "my documents" and retire the server. Took just over a year for us and worked fine.

We solved it by beefing up applocker and our AV running in MD5 hash hardened mode. We dont hide or restrict the drives any longer and pupils can try to run EXE from their USB drives, temp drives, local %appdata% if they wish (i.e. in places that they had access to previously) but cannot. Yes this is a pain when you want to install new software but a necessary one. Why run applocker AND AV? This is when visual studio becomes an issue. Some of our pupils are allowed to compile programs of their own creation for their computer science A level. AV picks up on these compiled programs and kills them. Applocker can be tuned to run with VS. If you want to run various applocker profiles, be aware that whilst you can add on applocker restrictions, removing them is a different issue. You need to clear registry keys if you wish to remove applocker restrictions. I cannot remember what they are right now but it is an easy google.

Its that time of year where im archiving old teams and starting our fresh academic year. I was going to do the same as I always did - use my PS scripts to grab from isams API and powershell them to SDS. This has worked just fine for the previous couple of years. However, I see that isams now support oneroster (albeit for a chargeable fee). Has anyone on here shelled out for the isams oneroster addon? Does it work? Hiccups? I have asked for a quote from isams for enabling the oneroster API (if the quote is silly then it will be a no go anyway). I must admit, i'm a little peeved (but not surprised) that it is a second chargeable API (given that we already pay for the isams API)

just remember that removing an applocker policy does not necessarily affect existing profiles. You may need to head to https://en.it-pirate.eu/windows-10-applocker-policies-still-affect-disabling-service/ and look at "option 3" However, applocker is good and once you get used to it, it can be powerful.

Ive just twigged. You are an SPLA. Ignore all I have said and you are correct. SPLA play by different rules, I have limited info on SPLA im afraid but it does look to be more expensive than us "on premises" people. 16 core packs are not available for SPLA and the licensing is per CPU core with 1 CPU and 8 core minimum pricing. The kicker is the OSE not counting! https://wiki.msp.exchange/sales/spla_windows_server_licensing shows it gets expensive quickly on STD

Why do you say that? Sometimes you dont need any licensing. Say you want to run a linux PFSENSE firewall on a server. You arent familiar with VMware so prefer hyper-v. You can install hyper-v server core on a server licence free and keep your linux pfsense container in that if you wish - no windows licenses needed at all. This makes backup/restore an easy option on dissimilar hardware as you simply need the virtual switches setup the same. The instance can be added to veeam for a much cheaper cost (as it is a virtual load not physical) for example. VMware licence per cpu up to 32cores so you still need to count CPUs etc which is more cost efficient, 1x 32 core CPU or 2x 16 core or 4x8core? No idea but the MS model allows mix and match whereas VM is per CPU almost regardless of cores (32 is alot per CPU at the moment!). If you go over 32cores per CPU then you need extra CPU packs. Oracle also have a core factor and CPU calculator etc. It is quite common. If you want REALLY fun licensing then look at a hybrid SAP onsite and cloud with updates. That has the lot, per CPU, per core, per software, per management, per CAL and hosting transfer licensing. Granted it has been a while since I played with SAP but it wasnt fun. Most people just took the very expensive perpetual licence and upgraded when they really had to....

STD and DC come in two licensing options, 2core and 16core. It is cheaper for us to buy a 16 core than 8x2core. We use DC as we need features in DC. We have 4x servers in a cluster, each of the hosts have 2x 8 core CPUs (makes it easy). We happen to have 9 VMs in the cluster. Since these VMs can migrate to any of the hosts, I need to license each server to the MAXIMUM VMs i.e. 9. So for us, licensing is easy - 4x 16core DC licenses. Now, if I wanted to go STD then this again is easy. Per server I have 9 (read 10) VMs and 16 cores. So that would be 5x 16 core STD licences per host (each STD allows 2 OSE VMs), 4 hosts in my cluster so 20x 16 core STD licences for my site. I do not use the hosts for anything other than hyper-v so I do not need a licence for the host itself. If your hypervisor was a domain controller (!!) then you would need an additional 16 core STD licence for the host alone in addition to the 5x 16 core STD licences for the VMs - you are NOT ALLOWED to use the "host alone" licence in your VM calculation you see. Not that it is a good idea to use the hypervisor for anything else anyway! hypothesis time. If I had 4x 20 core servers in my cluster, then I would need 5x 16 core STD plus an additional 10x 2 core packs per server. This means a total of 20x 16 core STD licenses and 40x 2 core packs for my site. The link you posted to the calculator is wrong. Here is the MS page stating 2VMs per STD licence https://www.microsoft.com/en-us/windows-server/pricing (again providing the hypervisor is ONLY doing hypervisor tasks) In none of the above I have talked about CALS or external connector licences. Since we use EES desktop packs, these are taken care of. ECs depend on 3rd parties you have connecting to your network externally etc. Licensing is fun.

teams can be a pain but if you are familiar with powershell you can script an awful lot to upkeep and administer. We have been using teams and sharepoint for some time, staff are used to it and it seems to do everything that we like. The hardest part was getting people off zoom and onto teams for webcam and meetings, for some reason our teachers were reluctant to use teams for meetings (even though they use teams for classroom stuff anyway!)

It is usually 10, remember that if the host is not doing anything other then hyper-v then you get 2 VMs per std licence (otherwise you need to license the host AND have sufficient licences for the virtual load). Assuming your core counts are basic 16, for 5VMs (non clustered) you would need 3x std 16core worth licences. It gets funky if you are over the cores of course but you still need to license STD or DC per 16 core minimum anyway. we pay 299 for our DC and 47 for our STD (annually 16 core versions) so for us it would be 13VMs break even point between using 7x std 16 core licenses and 1x 16 core DC licence). I only run 9 VMs but need DC for stretch clustering though...

Just another note, be careful making manual checkpoints on exchange servers. Make sure that your checkpoints for your exchange (or SQL server) VM are set to "production only". Standard checkpoints will most likely barf exchange. You can set VMs to create automatic recovery points, this will auto create checkpoints in intervals. Im sure that hyper-v replication make their own snapshots so if this is failing then these snapshots might not be clearing up after themselves. VEEAM also creates checkpoints, as do most backup software packages. If these are interrupted (or run out of space) then they may not clean up after themselves.