Old Firmware on Switches

Our Enterprise Switches are now out of date and not supported anymore. Are you guys always taking care to have Enterprise Switches that are on the newest FIrmware or at least update the firmware when there is an urgent issue or are you investing the money rather in other things?

I mean if you have a datacenter you better care for it, but in our own environment, with a closed building, basically no guests or so, should we really care to upgrade the hardware?

EDIT: How would you rate the security on it? All management Interfaces are on a Management VLAN and not accessible from anyone except our Privileged Access VMs.

https://redd.it/1pi7hx9
@r_systemadmin

I Fucking hate Microsoft

Fuck Microsoft. They changed the design again for the main Office home page. You can’t even find the Admin option anymore. Now you have to click on “Apps” first, and then you can pick the Admin option and pin it to the Office apps menu. Who designed this page? SMH. I’ve received so many tickets from users just trying to figure out how to open the apps from the main Office page. This Copilot thing really ruined everything, and now they’ve made this new change on top of it. Please, keep the Admin section separate from the applications. As admins, we should have a dedicated option under the apps. This whole design is so messed up — I hate it.

https://redd.it/1pib3rh
@r_systemadmin

Fibre channel vs iSCSI

A bit of an informal straw poll...

In my first job managing a datacenter for a medium business in the UK, and they have (before my joining) decided that they needed a separate storage network, using a pair of Brocade DS6520, connected to a Powerstore 3000T.

Being relatively green to datacenter infrastructure, Ive never actually seen this until now. Always dealt with collapsed core-type architecture, where SAN and LAN are over the same switches.

What's weirder, in my opinion, is the choice to have gone for storage switches that (currently) top out at 16Gbe per interface, while having 25Gbe on the LAN. We're currently hosting just about 200 VMs... If I was here circa 6 months sooner, I'd have pushed for iSCSI all the way.

Would love to hear stories, opinions etc.

https://redd.it/1pi9syf
@r_systemadmin

Kinda losing motivation to get into sysadmin

Just to be clear - SysAdmin is my end goal. I am applying for helpdesk/tier 1, 2 only. I have only applied for 1 junior system admin role and I had an interview for that. It's the only interview out of the hundreds of other helpdesk/tier 1,2 jobs I've had. This post is more of a help from you guys that are sysadmins and have been where I am do give me some advice or help.

Im 42. Been an industrial cleaner/team leader for 20 years. Decided to get into IT as thats what I wanted to do when I was young. Started my journey like 6-7 months ago now. Passed conptia tech+, a+ and networo+. Built a home lab. Learnt powershell, sql, excel, windows server, Linux server. I have a m365 business account and have added a few phones and vms.

I just can't get an entry level job at all. Ive had one interview and that was for a junior system admin and the interview went great and they were so close to choosing me but someone who they interviewed dead last had like 10 years it experience and because ive got 0 it was a no brainer.

I apply for so many jobs and only had 1 interview and that was only because my friend works at the company. The more I look at jobs and what they expect you to know is just putting me off and I just keep thinking if giving up and sticking to what I know even though I hate it now. Its mainly previous experience they are looking for

Any advice?

https://redd.it/1pih10y
@r_systemadmin

Hero Complex

Stop trying to be a hero.

Stop replying to every email within 20 seconds.

The best IT guys don’t try to be hero’s every fucking minute of the day.

You don’t need to be a hero and prove yourself even though you’re a classic imposter.

Silence. Patience. Refrain.

Actions speak louder than email replies inferring how great you are.




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

Fortinet - New Auth Bypass CVE for fortiOS 7.x FG-IR-25-647

Didn't see a thread about it yet but looks like all but the latest pretty much of all of the 7.x builds but the latest are effected https://www.fortiguard.com/psirt/FG-IR-25-647 as well as fortiweb/fortiproxy :/ Unclear if trusted hosts would prevent abuse, would think it would but since it's related to forticloud not 100% clear, just thought I'd post for awareness

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

Anyone else been force promoted?

I have been in IT for about 10 years now. I have been at the same company the whole time. The company wants me to step into a cyber security director role against my will lol. It feels like I live in a clown world sometimes. The impostor syndrome is real. I have been an soc analyst for 2 years....

I absolutely want nothing to do with managing people. Systems are much easier in my mind. So I am curious is it worth leaving a company that is forcing a promotion that I dont want? Important to add they have not delivered any raise yet. They also havent gotten that kind of work out of me yet because I won't do the work without the pay. Supposedly the money is on the way.

Supporting a few hundred servers and about 1500 endpoints.

Anyone else experience this or something similar? How did you handle it? If the answer is leave I am willing to I just love the people I work with and thats hard to find.

I do well on my own. I dont like to be stuck between my friends and top management. Translating that mess = a monkey humping a football!

I feel like maintaining my peace at this point is a more intelligent move, or maybe I should stop being a little bitch and "sack up" as they say? Embrace the suffering 🤷‍♂️.

Let's say I do stay, I would be managing two security team members two analysts and one engineer at some point. How much of a salary should I ask for? Thanks reddit mob in advance!

https://redd.it/1pim1de
@r_systemadmin

How do you handle certified destruction of HDDs/SSDs during large fleet refreshes?

In large-scale replacement scenarios, I keep seeing three recurring paths: NIST 800-88 overwrite for HDDs (one pass + verification), crypto-erase for SSDs where the controller supports it, and, when it doesn’t, physical destruction with controlled particle size. What mattered for us was having serial-to-device mapping before and after, a verifiable chain of custody, and reports that can go straight to auditors without extra translation.

For big batches we used E-Waste Squad specifically for the operational side: uniformed team and tamper seals at pickup, tight per-serial inventory, destruction certificates delivered within 24 hours, and reports that include serial matching plus timestamps for each stage. It also helped that their processes align with R2v3, ISO 14001, NAID AAA, and NIST 800-88-documented erasure, which cut down audit friction.

What do you require in the SOW when you outsource ITAD: on-site witness, photo/video of shredding, sub-24h SLA for certificates, CSV/JSON serial exports, or even on-site destruction for certain media?

https://redd.it/1pih7nd
@r_systemadmin

I now understand why other IT teams hate service desk

I started on a service desk, moved my way to L2&3 support then now to where I am in cyber security and while on service desk never really understood the animosity other people had for SD, I now really do! Whether it is the rambling "documentation", no troubleshooting or just lack of screenshots forcing me to chase up with the end user rather than actually fix the problem.


The issue is that while there are some amazing people working on it the majority are terrible. Something I forget is that most decent support people move out of SD as fast as possible so that the remaining are just shite.


Don't say "we did some troubleshooting" then not document what you actually did, and for the love of christ I'd take a blurry screenshot or even you taking a pic of the screen with your phone over nothing at all.


\- signed frustrated AF support person

https://redd.it/1pioxb2
@r_systemadmin

At some point in the past 10 years, configuration management went from open-source, to mostly paid/gatekept solutions...

I've been somewhat behind on employing configuration management software to standardize VMs: its only recently I have a stable enough environment to attempt this on again. That being said, the landscape is... changed...


* Salt's still around, but it's owned by VMWare, now Broadcom. Given Broadcom's behavior of late, I am weary of trying Salt again without running into some future license/legal demand.
* Perforce owns Puppet now: If you have less than 25 nodes, you're good, else expect to pay otherwise.
* Chef is now owned by some AI-focused firm: there appears to be a free version for non-commercial use, but the listed OS support is somewhat out-of-date.
* There's Rudder: it has a free tier, but it doesn't include Windows systems for endpoints.
* There's Terraform from HashiCorp, now owned by IBM: not really suited for my use case, but an option for others with "fleets" of systems.
* It looks like technically you can use Ansible (owned by RedHat, who's also owned by IBM) without a paid plan? Just need to be semi-proficient in Python.
* The one "truly free" option I found is Capistrano: requires some Ruby knowledge but appears to work for hosted application deployment; not sure about state-enforcement.



Right now, I have queries out to Perforce and Rudder for my small-scale environment, else I might forge ahead with an Ansible deployment. Otherwise, the purpose of this post is to let folks know what I found, and maybe find out if there are newer options not on my radar.

https://redd.it/1pip1ui
@r_systemadmin

Auditor asking for access review evidence we never recorded

We’re going through our SOC 2 renewal and the auditor is asking for evidence for everything (2024) like access reviews, onboarding/offboarding everything

Problem is this:
No one stored anything we don't have any screenshots or logs. The guy who owned security left six months ago and apparently he didn't document and keep track of everything
Now leadership is asking me to ‘recreate’ what happened last year (in my head I think it's impossible but I don't wanna give an answer without being 100% sure)
What do you suggest me to do?

https://redd.it/1pir9oi
@r_systemadmin

So tired of running into C-Levels who think Cloud/SAAS and Outsourcing are the answer to everything.

I’m so tired of having to change jobs every one to three years because a new CIO or CEO comes in and immediately decides, “Let’s move everything to the cloud or to SaaS, and then we can outsource whatever little in-house work is left.” They act as if we’re supposed to be cool with it—or even excited—that our jobs will disappear in a few months.

I see this pattern at every corporation I join.
How do others handle what feels like a constant, never-ending issue?

https://redd.it/1pisx29
@r_systemadmin

Microsoft 365 issues in Australia again?

Outlook isn't syncing, can't access admin centre from any internet connection.

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

I am in Remote Desktop Hell

I am two months into a new System Admin position and things are going pretty well overall, except for the Remote Desktop environment. I’m reaching out here as a last-ditch effort and hoping to draw on some of y’all’s experience.

Basically, for the last several years the RDS environment has been dealing with a whole range of problems. Users get profile-loading errors, sometimes they connect and just get a black screen, and most frustratingly there are random disconnects that seem to hit without any real pattern. Thin clients especially will drop the RDP session after being logged in for about two minutes. Event Viewer on the hosts hasn’t been very helpful, but on the client side I’m consistently seeing a TCP socket error. At this point I feel like I live in Event Viewer and I’m constantly chasing my tail with nothing ever actually improving the connection.

It is a Windows Server 2022 RDS environment supporting under 1000 users.

What I Have Tried:
I’ve made a number of changes through Group Policy, including adjusting session timeouts, security settings, and RDP encryption levels. I’ve combed through the logs on both the hosts and the clients repeatedly trying to correlate disconnects with any specific event. I’ve checked the health of the broker, verified certificates, and confirmed licensing is functioning. I have even captured packets in Wireshark to try and see what the disconnects look like on the wire, but nothing has clearly pointed to a single root cause. Despite all of this effort, (This really has consumed my last couple of weeks) I have seen minor improvement on the profile errors and basically no improvement on the disconnects.



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

Anyone actually pulling Entra risk/NHI signals into their SASE console yet?

Trying to get real Entra identity health (user risk, signIn anomalies, NHI scores, leaky token alerts, etc.) to show up natively in our SASE dashboard (Cato, Netskope, Zscaler, whatever) instead of just basic "user authenticated" events.

Docs only talk about the standard Entra IDP connector. Nothing about the deeper risk telemetry or identity protection feed.
Has anyone cracked this in production? Graph API polling? SCIM hack? Direct feed from Defender for Identity?

Real experiences only, please. Thanks. (Im already convinced that it might not be possible but still need to see if by any chance there is any possibility?



https://redd.it/1pizuc9
@r_systemadmin

Print drivers

Hi All,

I updated the driver for our canon copiers on our Windows print server. Most of our devices have received the new driver and are working fine. However, it seems some of our devices are holding on to the old driver, and only obtaining basic settings from the copier. (only A4 size paper, no hole punch, etc.) Is there a way I can force this driver to update on the end user side? I've been having to manually grab the device, stop the spooler, delete the driver, start the spooler, and reboot. I'm not sure how many of these are broken like this in the wild, so i'd like to find a better method.

https://redd.it/1pj5fxk
@r_systemadmin

What do you do when a vendor screws up?

I work for a small local government org. We have to do some minor broadcasting of meetings which I will admit is a bit out of my realm. We recently had some issues during a meeting and found out that they were due to a switch losing a config after a power outage that a vendor forgot to save... While we have a good relationship with them, it's hard to swallow paying a service fee which will probably only be about $1k maybe even less. Would you hammer them to cover it or let it slide?

https://redd.it/1pj2wyo
@r_systemadmin

VMware

Any of you guys being f-ed over by your VMware renewal this year? Ours went from 11k last year to 65k this year.

https://redd.it/1pj7z68
@r_systemadmin

Windows Server 2019 is rebooting itself when updates are applied, even though I have it turned off

I have a Windows 2019 Server hosted on Azure that rebooted itself during the day yesterday which brought our production system down.

The message in the System Event Log is:

The process C:\Windows\system32\svchost.exe (MyServerSQL) has initiated the restart of computer MyServerSQL on behalf of user NT AUTHORITY\SYSTEM for the following reason: Operating System: Service pack (Planned)
Reason Code: 0x80020010
Shutdown Type: restart


We are a small company that doesn't have a sysadmin, and I'm a developer trying to manage these systems. I have turned off the "reboot after update" on the Windows Update Manager, but I've obviously missed something.

What can I do to ensure that this isn't rebooted unless I say so?

https://redd.it/1pj5boi
@r_systemadmin

So what software do folks use to run VMs these days?

Not bare metal hosting like Proxmox, but running VMs on Windows. My go-to used to be Virtualbox, but it's been awhile since I've messed with this and I wasn't sure if there was a better way.

Apologies if this is a dumb post, I just wanted to make sure I'm using the latest and greatest.

Thanks!

https://redd.it/1pjbn4h
@r_systemadmin

Value of VMware ESX-based knowledge?

How worthwhile is it to learn VMware ESX-based virtualization these days? How valuable is this knowledge today? I am considering purchasing a Udemy course on the subject. I am interested in virtualization, but so far I have only had experience with Proxmox.

https://redd.it/1pjdl8d
@r_systemadmin