@siringo said in windows based FREE imaging app:

hello.

i'm looking for a windows based (or anything that can be used easily by a non linuxite) FREE imaging app that can perform restores via a script / command line.

I have to reimage 50 new PCs.

in the past i've used free gui based apps, booted up, clicked here and there and the job was done. But i'm getting lazy and would just like to be able to boot up, click on script and walk away.

I could afford up to 10 bucks if the right thing can be found.

thanks everyone.

You try MDT?

@notverypunny said in windows based FREE imaging app:

@obsolesce said in windows based FREE imaging app:

@dashrender said in windows based FREE imaging app:

@obsolesce said in windows based FREE imaging app:

@dashrender said in windows based FREE imaging app:

they generally come with AV and other crap you don't want at purchase

Oh I see, that sucks. Are the company devices being bought from Walmart or something?

Seriously?

I order these from DCW. I haven't had a laptop not come with at least some third party AV in ages...

I suppose one of the reasons to not order Dell/HP, or at least not the default stuff.

Can't speak to HP, but with Dell, unless you get setup with their imaging program (you provide them with your desired stock image and it's $$$ from what I recall) they're sending you their stock OEM image with a significant amount of bloat-ware. In a corporate / enterprise setup consistency is king so it's normal that you want to reimage with something that's tested and known to play nice in your environment.

Business class devices shipping with trial anti-virus software that is well known to be much worse than the default Windows Defender? That alone is reason enough not to go with that manufacturer (still not a showstopper, as automation can fix that in later steps). If you need to touch a device before an end user gets it, you're wasting a ton of time and money. That's decades old procedures... having your IT department receive the device, reimage, configure, maintain images, and all the requirements that go along? That is a huge waste of resources.

Wouldn't you rather have a device sent directly from CDW to the end-user, without needing a special image, ready to go for the user and the work environment... managed, configured, secured, and compliant as part of the OOBE?

@hobbit666 said in Gaming PC Setup:

@obsolesce said in Gaming PC Setup:

@hobbit666 said in Gaming PC Setup:

@obsolesce said in Gaming PC Setup:

Personally, I'd not use HDD in any PC anymore, no matter the size. I'd choose to have a single 128gb SSD over 2 1TB spinners any day, and upgrade the SSDs when needed or even initially.

Not going to install much on 128gb now a day

I know, besides the point. I'd rather do what I can with a 128 vs 2tb spinner. Of course I'd immediately get something bigger than 128,but I was just making a point.

I know what your saying but this is what I'm working with

So would you say maybe just use the 2x 240gb ssd (Raid)
Going to keep one 2tb as she wants to get into youtube/twtich etc so will need some storage for video editing

If it was my setup, I'd use them as individual drives, havi g a and D:. If space starts to get low, then I'd install games on the other one. I don't know of any games that do not let you specify the location to install.

If you did raid, then you're stuck with 240gb. If you did raid 0 then your failure chance is double and all is gone. I personally don't mind 2 separate drives.

Its up to you to do how you like.

@hobbit666 said in Gaming PC Setup:

Need some advice on how to configure this guy for the kid.
ROG STRIX B450-F GAMING II
Amd Ryzen5 3600
16gb ram (will have to check speed)
RTX 3060 ventus 2x 12g oc

So the big question is this, I ordered this without HD's as I have x2 240gb ssd. But it came with x2 2tb.

Looking my initial thought was to use AMD StoreMI where you use the SSD as cache for the most frequently accessed stuff.

But would there be any benifit of make RAID 0 out of the pairs instead. Then maybe using the StoreMI and use the SSD Raid as cache for the HD Raid?

Or just pull some drives out and not use them

Personally, I'd not use HDD in any PC anymore, no matter the size. I'd choose to have a single 128gb SSD over 2 1TB spinners any day, and upgrade the SSDs when needed or even initially.

@dashrender said in Windows 11 auto upgrade?:

@voip_n00b said in Windows 11 auto upgrade?:

@dashrender said in Windows 11 auto upgrade?:

Plus I'm really trying to get away from local AD.

What's that have to do with MDT?

Perhaps nothing, though I've never heard of anyone using MDT outside of AD

It's used all the time without AD.

@pete-s said in What do you think about .app domain names?:

@scottalanmiller said in What do you think about .app domain names?:

If it is under the hood, why bother. If it isn't under the hood, I think customers get confused.

So you mean if it's customer facing it's better to stick to .com and there will be no confusion?

That is the only aspect that matters tbh, what people / customers think of it. All other aspects have zero impact.

@dashrender said in Wsus for remote vpn and on-premise users:

@obsolesce said in Wsus for remote vpn and on-premise users:

You can use Windows Update for Business. No need for WSUS.

Is there any type of reporting in that?

Yes, multiple methods of reporting... reporting out the ass.

Still using LastPass Families. Works well for us, no reason to switch to something else.

We keep it simple, vanilla PC through Autopilot if possible, or set up PC with work or school account. In either case, sign in with Okta account and Intune takes over... Config, compliance, policies, remediations, required apps, settings, etc. Typically from opening box to end user working in less than 15 minutes without IT ever having touched the device at all.

@d-cunnings said in User Profile migration Problem AAD -> AD:

Customer pulling in smaller firm running Windows clean Azure.

I am to get those users off their Azure and onto the On-prem domain and have been given the task to move not only their data but also their current user account experience.

Going through everything I could find over the weekend I get to the point where I conclude that there is no way to do this.

-There doesn't even seem to be a way to link a local or domain profile to an Azure account?

Has anyone of you done this?

Maybe just get rid of computers and go back to pencil and paper?

@xavierdelaraunt said in User Profile handling anno 2022:

I have quite a lot of settings I cannot get into the default settings in any way like that.

I have not ran into any policies, settings, etc that could not be don't through Intune. It would be very hard for me to think of something that I can't do to a device related to those through Intune.

If you are referring to user profile migrations across different devices (old device > new device), most of that can be resolved by better systemic practices prior. That can be a shitty experience regardless of MDM or device management system.

Just to make it clear, I'm using rclone sync to do the job on a headless Ubuntu container.

Revisiting this now two years later, as it's been that long since I last synchronized my personal OneDrive photo and video archives to Wasabi.

Most everything is great, except when doing a few --dry-run tests on some already-synchronized source/destination remotes, I noticed a few thousand files in total among several OneDrive archives result in a potential re-sync. Not a huge deal, except it's a slow process to sync.

The files themself didn't change. I verified from both the source and backup that the CRC is the same on a bunch of to-be re-synced files, but perhaps the modification time changed at the source due to other syncs or who knows.

I would like to force RClone to use checksum only when synchronizing. Is that possible with an rclone sync between OneDrive and Wasabi?

@Pete-S said in Production KVM server "hardening"?:

I'm thinking about running pure KVM on debian for virtualization hosts. Not Proxmox. There will be no GUI on the servers, no web interface, only ssh for management.

Do I need to do anything special to lock down the security?

I've never used KVM in production, only on my desktop and then I've had virt-manager as well as tools like virtsh. So I don't really know what is required for a pure KVM server to be as "secure" as proxmox, xcp-ng or whatever.

Keep the OS and everything updated. Keep drivers updated. Keep firmware updated. Use only key-based auth for SSH, add only specific devices to authorized_keys file. Ensure firewall configured well. Set up log alerts for access.

@JaredBusch I did not use an elevated ISE window:

And it worked for me:

@Fredtx said in 1 large disk or 2 smaller disks for a file server?:

I'm creating a new file server, which will be a VM. The source (original file server) has 1.7TB of used storage. 641GB of that is Marketing (mainly videos), and the rest is Engineering (cad files), User folders (docs,etc), and miscellaneous folders.

Would it be better for me to create 2 virtual disks on the target (new file server), and give the Marketing team their own disk? Or should I just move everything over to 1 big virtual disk?

2 virtual disks versus 1 virtual disk would only matter if you need to do some kind of QoS at the vdisk level. Also, depending on how you do shares and your backup infra may matter, or maybe not.

@DustinB3403 said in Query Regsitry using a Variable:

@Obsolesce Yeah I got that far along, what I need to pull is a specific string from the output.

I have a few other ways that I was manipulating the string, like writing the entire output to a file and then pulling the 23rd line (for example) but that literally gets everything on that line.

When all I want is the InstallDate

Get-ChildItem -Path 'HKLM:\SOFTWARE\WOW6432Node\Microsoft\Windows\CurrentVersion\Uninstall' | Get-ItemProperty | Where-Object -Property DisplayName -EQ "Microsoft Edge" | Select-Object -Property InstallDate

$InstallDate = Get-ChildItem -Path 'HKLM:\SOFTWARE\WOW6432Node\Microsoft\Windows\CurrentVersion\Uninstall' | Get-ItemProperty | Where-Object -Property DisplayName -EQ "Microsoft Edge" | Select-Object -Property InstallDate
$InstallDate
$InstallDate.InstallDate

I did this in the past, but not sure how relevant it is for you here?

https://mangolassi.it/topic/17916/ubuntu-setup-for-steam-play-and-lutris-gaming-with-nvidia-graphics?_=1670778770294

I've just been gaming on Windows lately, bottom line is it's just easier and less time consuming for some of the games I play. So it's dependant on that.

@BraswellJay said in Word to PDF converter ...:

@Obsolesce said in Word to PDF converter ...:

@BraswellJay you don't need one with word. In Word, just export as pdf.

I can't do that in bulk that way though, can I?

My user may have 50 or so word files and she wants to convert them all to separate PDFs without having to open each one individually in word.

I thought there may be a tool that would just let her bulk select all and convert them all at once.

You can do it with PowerShell. Make a simple script for the user to run.

@Pete-S said in GitLab Now Integrates VS Code Into the Browser:

@scottalanmiller said in GitLab Now Integrates VS Code Into the Browser:

This is a really cool update to GitLab. Microsoft's VS Code is now the web IDE used online in GitLab. So you can use VS Code without needing to install it. This is the coolest!!

That's interesting.

Editing on the webserver breaks the idea of how git is suppose to work though. Basically makes it a central version control and repository, instead of a distributed one.

Good to know that it exists though!

Of course, it totally depends on what you're working on. Now that that's out of the way...

You can still do a lot the same as far as branching and PRs. If you're not compiling or testing things locally or need local resources, then there's not much other benefit doing it on your local PC versus some more simpler editing directly, besides some efficiency factors. But that other stuff should be done automatically anyways when you (for example, create a PR) via automation / pipelines.

If I need to fix something real quick and don't have my local environment ready or for whatever reason, it seems nice to not have the typical shitty editor like GitHub has, and I can easily create a new branch and make the changes, create a PR, and get it merged in in the end. It's not a bad thing to be able to do it from the browser, the same as you would from your local environment in VSCode (if what you are doing works out that way). All the testing, cleanup/linting, security checks, building, etc. should be kicked off automatically anyways no matter from where your changes come from.

@Pete-S said in Proxmox hates security:

KVM by default for instance is managed by libvirt and by default doesn't open any tcp ports at all. That gives the administrator the option to decide what level of security versus convenience they want.

Remotely using Virt-manager for example, via remote SSH connection.