A coworker died yesterday. There's an important lesson that some need to hear.

Yesterday, the company sent out an email announcing that Bill, who had been with the company 20 years, had passed away suddenly. Everyone liked Bill. He was one of those genuinely good people.

Today (the very next day), we had our annual catered lunch. The entire staff was there. And not a single word was said about Bill. No moment of silence. No acknowledgment at all. For 20 years this dude had been there, and everyone just went on with their day, eating and chatting like nothing had happened. Like the guy never existed.

We spend 40 hours a week with our coworkers, that’s more time than many of us spend with our own families. And I want to be clear: I like where I work. It’s a great place. The pay is good, and the people are solid. Truly.

But this really drove something home for me. As the holidays come up, management LOVES to pile on projects for IT on these days when everyone is gone, system upgrades, network changes, “quick” maintenance windows. And I say fuck. that. noise.

No one will remember you swapping a phone system on New Year’s or replacing a UPS on Christmas Eve. But your family will sure as shit remember you not being there. And if you die suddenly on a Monday, odds are your coworkers will have moved on by Wednesday.

Say no. Stand your ground. Protect your time. Be with your family.

https://redd.it/1pe971k
@r_systemadmin

What tools did you use to automate onboarding?

Onboarding for us, and some of you I’m sure, is a very annoying, labor-intensive process, all because there is very little automation.

For the past year as a back-burner side project, I’ve been gathering requirements from each department that touches the new hire process in any way.

At this point, I’m just blind to my options because I’ve never done this before in my career. In my research, I am considering Power Automate and set up as may triggers and dependencies as I can, and leave certain things to manual process, but other than that, I have no direction or knowledge of the COTS solutions out there.

What do you do for onboarding? I’m not looking for what happens during your personal business process. I’m asking specifically about what tools and solutions worked for you in your org? Hoping to get some traction and places to look.

https://redd.it/1pe09dz
@r_systemadmin

Sys admin sucks update

Prev post

I was going to post this update sooner as I recently walked out one day due to harrassment.

This rant will include things that I have heard or that a colleague has heard.

storage of plaintext passwords for crucial staff members

you require AD to run a simulated phishing campaign through email

Scripting is not allowed as it'll automate us out of a job. "Scripting isn't allowed because there's no way to know if it worked." (I noscript anyways)

It isn't possible to have a netlogon noscript not include their password in plaintext

"You can't be expecting these changes to happen right away it takes time" you've been working on AD for how long? there is no progress.

in my interpretation, privacy law violations. (plaintext passwords)

no longer required to use 2/3 of the programs I described in my last post

So far I've heard an IT guy at another organization receive more on the job training from the sysadmin than I have (not that I want to learn anything from this guy anyways)

One of my colleagues set up AD for one of our departments and the sysadmin convinced a higher up that we "weren't ready" for AD and then he got paid overtime to delete the entire server and rebuild it from scratch with local accounts.

There was a day where he had a 30 minute rant about AI hacking your pc and uploading everything if you use it once (chatgpt, copilot)

"Hackers are in the cloud, so we don't recommend storing anything there."

If you get "hacked" through your email on a work laptop you have to let him wipe your personal phone if you at any point logged into your email on your phone or if you even use teams.

He does not wipe work laptops when they've been infected, just runs virus scans.

I'm just collecting a paycheck at this point and have mentally checked out. There is still so much more but this is more of the current stuff.

https://redd.it/1pe6291
@r_systemadmin

Replace Server 2008 DC with Server 2025?

EDIT: Great news! We convinced the customer to terminate the old domain with extreme prejudice and just create a new one. Every single employee was a domain admin on the old domain and there were tons of other problems with it. Win-win.

Am I fucked? Everything I'm seeing says I literally have to install a temporary 2012 server first.

The 2025 server won't promote because the forest functional level is too low. The 2008 functional level says it is as high as it can be.

Do I really have to do a temporary server?

edit: because I have a tiny amount of pride, this is a customer. I've done some stupid shit, but I take zero responsibility for having a 17 year old DC.

https://redd.it/1pe8955
@r_systemadmin

Adobe Acrobat Pro 2020 end of support Nov 30 – new signed PDFs already broken, need cheap 2024 perpetual / term license alternatives fast

Inbox on Dec 2nd ruined my week.

380 seats of legit Acrobat Pro 2020 we bought outright back in 2020.
Adobe email hits: “EOS Nov 30 2025, your installs no longer validate ISO 32000-2 signatures. New signed docs already show validation errors.”

Every single contract or invoice we get now opens with the giant yellow “SIGNATURE VALIDITY UNKNOWN” banner. Legal is losing their minds, compliance audits looming.

Adobe quote to stay legal:

$72k one-time for 2024 3-year term licenses
or $90k+ yearly subnoscription forever

Foxit pilot was a disaster, redaction sucks.

Anyone found a real volume reseller still moving cheap Acrobat Pro 2024 term / perpetual licenses with proper CLP paperwork? Or are we all just getting forced into the subnoscription hell at this point?

https://redd.it/1pehlel
@r_systemadmin

NSF I-Corps research: What are the biggest pain points in managing GPU clusters or thermal issues in server rooms?

I’m an engineering student at Purdue doing NSF I-Corps interviews.

If you work with GPU clusters, HPC, ML training infrastructure, small server rooms, or on-prem racks, what are the most frustrating issues you deal with?
Specifically interested in:

• hotspots or poor airflow
• unpredictable thermal throttling
• lack of granular inlet/outlet temperature visibility
• GPU utilization drops
• scheduling or queueing inefficiencies
• cooling that doesn’t match dynamic workload changes
• failures you only catch reactively

What’s the real bottleneck that wastes time, performance, or money?

https://redd.it/1pem1yl
@r_systemadmin

Does anyone else feel like they can't predict how long anything will take anymore?

And how are you dealing with this in terms of setting expectations/SLAs with clients or end-users and not constantly feeling like you can't make even minor guarantees/promises about providing a reasonable level of service?

I keep having situations where the same tasks, projects or issues vary wildly in their turnaround/TTR simply due to stupid, unpredictable, inexplicable sh*t like:

Progress bars getting hung for no reason or the same compute tasks on the same hardware just magically varying in completion times because the devil inside the silicon knows you're in a rush so fuck you and your weekend plans
Downloads taking way longer to complete than normal
Servers being unresponsive/busier than usual, again for no obvious reason
Random service provider/SaaS outages or service incidents that prevent timely access to urgently-needed resources and platforms
Never-before-seen error messages, bugs or crashes in the middle of something you've completed 1,000 times before without issue
Major players like Microsoft/Amazon constantly making rug-pull-stealth-changes to major parts of their ecosystems, core services and UIs that you never see coming until you're frantically trying to do something you've confidently done many times before (like I don't know... logging into a portal) and now you're confidently flailing aimlessly until you submit to relearning their processes for the 1,000th time.

It's these kind of side-tracking bullsh*t detours in the middle of already insane workloads and razor-thin deadlines that I can never find a good workaround/Plan B for.

Am I supposed to be operating triple redundant workflows and processes like I'm flying an airliner or something?

Or is the answer supposed to be that I start every single planned piece of work days in advance of when I normally do, even though that is obviously impossible most of the time?

I feel like I just end up delivering everything a day late and a dollar short because of circumstances that are largely out of my control but that still reflect poorly on me because clients and end-users don't realize all of the complicated, moving pieces at play in performing task X or fixing problem Y.

https://redd.it/1pepjmc
@r_systemadmin

Cloudflare down again?

Can't reach anything cloudflare-hosted from Sweden right now.

https://redd.it/1peq658
@r_systemadmin

Weekly 'I made a useful thing' Thread - December 05, 2025

There is a great deal of user-generated content out there, from noscripts and software to tutorials and videos, but we've generally tried to keep that off of the front page due to the volume and as a result of community feedback. There's also a great deal of content out there that violates our advertising/promotion rule, from noscripts and software to tutorials and videos.

We have received a number of requests for exemptions to the rule, and rather than allowing the front page to get consumed, we thought we'd try a weekly thread that allows for that kind of content. We don't have a catchy name for it yet, so please let us know if you have any ideas!

In this thread, feel free to show us your pet project, YouTube videos, blog posts, or whatever else you may have and share it with the community. Commercial advertisements, affiliate links, or links that appear to be monetization-grabs will still be removed.

https://redd.it/1per9qg
@r_systemadmin

Cloudflare is down again. Two outages in two weeks. Anyone else concerned about the dependency chain here?

Cloudflare is having issues again today and it feels like a repeat of what happened two weeks ago. Same pattern. Perplexity stalls, Claude stalls, auth flows stop responding, and random internal tools start throwing cryptic errors until someone checks the status page.

Two outages in this short a window really highlight how much of our infra hangs off a single external point. It is not just websites that stop loading. It is SSO, API calls, AI platforms, monitoring dashboards and even internal automations that have nothing to do with Cloudflare on paper.

I am curious what the sysadmin community thinks. Is this just the reality of relying on massive edge providers, or are we getting too comfortable with architectural bottlenecks that fail in unpredictable ways? Are any of you actually planning around this or is it just accepted cost of doing business now?

https://redd.it/1peqk0l
@r_systemadmin

Which is the Best EML to PST Converter For 2026?

I am looking for reliable and best EML to PST Converter, which can easily convert bulk EML files to PST as I am in trading firm so get wide range of EML files from small to large sized EML file. I want to access or import EML files to Outlook PST format. That's why, I need a reliable solution.

My recommendation are simple, the tool can easily convert large number of EML files into PST format without losing any data. I have thousands of EML files collected from different email clients so it should be able to handle batch conversion smoothly.

The tool should also maintains the original folder structure, preserves email metadata, and keeps attachments intact throughout the process. The tool must offer simple user-friendly interface and support to Windows and Outlook versions.

Recommend a professional EML to PST converter for bulk migration

https://redd.it/1pes12h
@r_systemadmin

I think its time to look Cloudflare alternatives.

The Cloudflare centralization risk is no longer theoretical. It’s time to talk about "Eggs in One Basket."
We are watching half the internet go dark again today (Dec 5), barely a few weeks after the November 18th outage.
20% of the web went down because of a single bug in their Bot Management logic that "failed closed." When a single vendor's feature update can inadvertently wipe out that much traffic globally, we have reached a dangerous level of centralization.
we talk about high availability and redundancy for our own stacks, yet we are routing everything through a single proxy that is becoming a SPOF for the entire internet.


https://redd.it/1per48w
@r_systemadmin

Cloudflare down again?

Is Cloudflare down again? Started receiving a lot of "500 Internal Server Error cloudflare" error messages now on various websites.

https://redd.it/1peqegk
@r_systemadmin

Cloudflare outage now in status page

From https://www.cloudflarestatus.com/

'These issues do not affect the serving of cached files via the Cloudflare CDN'


... I think they do

EDIT: That line has already been removed from the status page

https://redd.it/1peqblt
@r_systemadmin

blue tally for 120-laptop youth nonprofit?

We are a small after-school youth nonprofit with about 12 staff, 160, 180 teens per semester, and roughly 120 laptops plus some tablets and a handful of desktops.

Right now all device tracking is in one Google Sheet I inherited. It is… messy. I have been looking at moving to an actual IT asset management tool instead of spreadsheets. BlueTally came up a lot in searches, seems focused on hardware, talks about lifecycle logs, integrations with intune/jamf, SOC 2, etc. But most of their case studies are big companies or higher ed, not tiny nonprofits.

Given our scale (120-ish laptops, maybe up to 150 in a few years, no full-time IT), is a dedicated tool like this worth the money and overhead, or is it total overkill and I should just fix the spreadsheet and processes?

https://redd.it/1pewdat
@r_systemadmin

Trying to prevent them shooting themselves in the .... foot

Background ... I work for an MSP. This particular client has a PUBLICLY VISIBLE service that I manage behind a proxy. The proxy has been having issues for the last couple of weeks which is causing availability issues in my application. The client has decided to pull the service off of the proxy. In other words, they want me to put a Windows-based server bare to the internet with no proxy, no edge scanning, no nothing .... just basic firewalls.

Now, I recognize that the platform is THEIR property and they can do whatever they want with it. But I also think that the biggest thing they pay me for is expertise to protect them. And so I feel like I have a moral obligation to just tell them no. I'm the one who has to turn the wrenches, so to speak, to make this happen. I could just flatly refuse to do it. Or maybe just demand it in writing and suck it up.

IN short ... client asks you to do something INCREDIBLY stupid. Do you cheerfully pick up the ticket and work it without complaint? Do just do your best to warn them and then work it? Or do you tell them "I don't want my name associated with something this stupid."?

https://redd.it/1pexfa9
@r_systemadmin

Sooo, what brand memory to buy now?

Buying Crucial RAM has been the default for me for many years. I never even looked at any other brand.

Now that Crucial is gone, what are you guys doing for memory upgrades? I realize this is a difficult time now with the DRAM shortage and price hikes. But assuming normal market dynamics (which will hopefully return), who do you trust for DRAM?

https://redd.it/1pf158c
@r_systemadmin

CIO and CTO want Office icons back on desktop again....

Way back in the day the Microsoft Office Pro installer had the ability to create shortcuts for the Office programs on the desktop as part of the installation by using the /admin switch and then configuring the option to do so.

We have not done that in some time now, obviously, since the Office installer is C2R and not MSI and apparently there is no supported way to do this with the published configuration information for the XML file during the installation of Office.

The CTO and CIO now want the icons back on the desktop again. I am hoping that I am just missing some obscure entry in the Office deployment tool documentation, but short of that am I looking at noscripting this out with PowerShell and then keeping up with asinine changes to directory struct for Office when and if Microsoft makes some?




https://redd.it/1pf2qn0
@r_systemadmin

Restrict ChatGPT access to company plan only

We allow a small group of employees to access paid ChatGPT Business. How do we enforce sign in / ensure that they do not log out of the company accounts and start using their personal plans instead?



https://redd.it/1pf08g7
@r_systemadmin

I tried read only Fridays today

Decided to just read through emails and see if anything was an emergency. In the mean time I focused on certification training and testing out some things. Was absolutely glorious.

https://redd.it/1pf6o7f
@r_systemadmin

New CIO without technical background relying on consultant

We've got a new CIO with a Finance background and the first thing they've done is brought in an architect to assess everything and create a roadmap for us.

They were an internal hire and have never worked in IT before, so they've needed almost everything explained to them between the IT team and the consultant. I can see the Finance experience coming in handy when trying to optimise costs but it still seems odd to me - bringing someone in that needs to outsource most of the relevant technical skills? Is this normal?



https://redd.it/1pf7awb
@r_systemadmin