I support an office that used to think rebooting computers was bad luck. Whats the weirdest bad behavior you have had to cure on an office wide level?

Posted by simAlity@reddit | sysadmin | View on Reddit | 320 comments

apparently there had been a day when 2-3 computers had crashed after reboot. One of them belonged to the administrative assistant that pretty much managed the office. Word got around that restarting computers was bad luck. Group policy here was absolutely horrendous. Automatic updates were blocked. Machines were 2-5 years out of date. Several hasn't been restarted in 6 months. I ended up doing in place updates to Windows 10 21H2, implementing automatic updates using vendor software and mandating twice monthly restarts. Now their superstition is just a unhappy memory.

Reply to Post

Reply

320 Comments

[-]

Secret_Account07@reddit

Uhhh how do you patch windows without rebooting?

Reply

[-]

Sure-Squirrel8384@reddit

hah, patching breaks things. "We don't patch!"

Reply

[-]

Secret_Account07@reddit

My team works with patching thousands of windows servers. The times servers go a longggg time without patches it’s like something happens. I can’t existing it but it’s like MS doesn’t test new monthly patches on a server that hasn’t been patched in 3 years. Weird stuff happens and it can have issues while patching or months later. I have no proof that applying X amount of monthly patches is detrimental to a system, but I stg it’s true. This may not be an issue with newer OSs as at have cumulatives now, but my many years of 2008/2012 (r2) patching knows it to be true lol Luckily it’s an issue of the past for our org, we get a ticket every single time a server fails for a month, but it didn’t used to be that way and I know a lot of orgs are wayyy out of compliance

Reply

[-]

Sure-Squirrel8384@reddit

Do you even English? Maybe proofread. "I can’t existing it " " but I stg it’s true."

Reply

[-]

Rude_Growth_9347@reddit

In our office, it's black for the VoIP phones and pink for the LAN cables. Every drop on the separated network has cables that match the jack, and even pretty little stickers of a phone and PC on the jacks themselves. Even with that, most people still fuck it up and then open numerous tickets about how the phone doesn't work and they can't get to the internet...... There's a reason that on my work PC, the number one search on Amazon is what I gently call the "User Training Tool". AKA a cattle prod. BOFH for ever! At least not in my building....

Reply

[-]

doofusdog@reddit

Wireless is faster than cabled One branch. Perpetuated by the manager.

Reply

[-]

crysisnotaverted@reddit

I have tested speeds up to 1.35 gigabits on WiFi 6E/7 using iperf3, so \*sometimes\* it might actually be true.

Reply

[-]

Injector22@reddit

But howany clients were connected? How far away were you from the AP, what's the ap density, what's the noise floor, etc, etc, etc. This is the issue with wireless, WO many variables could affect a test.

Reply

[-]

kanzenryu@reddit

If you connect ten clients it will be ten times faster, get with the program

Reply

[-]

Injector22@reddit

Well damn, I've been doing it wrong this whole time. I'm going to go remove all but one AP so we can get 100Gbps per client. I'll post the results shortly.

Reply

[-]

crysisnotaverted@reddit

This was just as a test, from wireless client to wireless client. Only two clients in that case. I put both clients 10 feet away from the AP. Noise floor was pretty low in the 6GHz band, as I believe this was about 2-3 years ago before widespread adoption. It was son a crowded office building. Now this is pretty far from lab grade conditions, but it was really just to sate my curiosity lol. I was tickled pink seeing above 1 gigabit fly through the air, coming from shitty Netgear WiFi B APs that used a PCMCIA card for TX/RX.

Reply

[-]

boli99@reddit

well, we tested it with one client and it was incredibly fast so then we filled the entire place with metal shelving and cupboards and deployed 250 mobile users /s

Reply

[-]

Smith6612@reddit

Oh yeah. I get 1.6-1.8Gbps on the WiFi 6E hardware at home. Speed is definitely not an issue on a proper wireless deployment. Many laptops still only come with 1Gbps Ethernet, and some super cheap ones 10/100 ports making the WiFi faster no matter what.

Reply

[-]

crysisnotaverted@reddit

I hear ya. The easiest way for me to get faster speeds on my 14 year old daily driver laptop was replacing the WiFi card with one with an Intel 6E AX210 chipset. Pretty crazy how much things have changed.

Reply

[-]

Smith6612@reddit

Surprised you got an AX210 into a 14 year old laptop. Did you have to get a conversion card? Most of the AX210s I've seen require a laptop with an M.2 slot.

Reply

[-]

crysisnotaverted@reddit

Yeah, most wireless cards use some form of M.2 with an A+E key on the pins. Some lovely people put a chipset on a card I could actually use though, I have the AX210 on a half-height Mini-PCIe! https://preview.redd.it/wmbbeql1njfg1.png?width=1000&format=png&auto=webp&s=94e092f69df2f03fed4400669b5c95e2dc00565f Definitely a strange little ugly-duckling. Shows that this technology generation isn't a total dead end.

Reply

[-]

Smith6612@reddit

That's awesome. I don't know how long I'll keep daily driving my current laptop, but I might toss in an AX210 or equivalent to replace the Realtek RTL8822BE on it (802.11ac Wave 2).

Reply

[-]

crysisnotaverted@reddit

Thanks! It's a worthy upgrade. Probably the cheapest upgrade possible with RAM and SSD prices right now. Only cost me $25 and I used the pre-existing laptop antennas.

Reply

[-]

Smith6612@reddit

Only thing I'd have to double check is whether the card will work properly with a first generation AMD Ryzen processor. Sometimes AMD Blocklists newer Intel cards from working limiting options to Realtek and MediaTek, and sometimes 6Ghz won't work because of the ACPI tables missing information needed for Location Based Regulatory to work in the card's firmware. Maybe the fact that HP refused to upgrade the Microcode on my laptop from RPi1.0.0.0 will be a blessing in disguise for Wi-Fi compatibility lol.

Reply

[-]

crysisnotaverted@reddit

There's DRM for this shit now? I only ask because the wireless card is attached to the PCIe lane of the laptop's Intel Ivy Bridge (3rd Gen) CPU. In my experience, the laptop didn't care at all about the card being a decade newer, the OS took care of it.

Reply

[-]

Smith6612@reddit

Yep. Due to FCC regulations in the US. 6Ghz is a bigger mess than 5Ghz.

Reply

[-]

pdp10@reddit

It actually isn't due to regulations, that's just a convenient excuse. Emissions on WiFi cards are handled by firmware loaded at runtime by the kernel, and the firmwares come from the chip vendors (Intel, Mediatek, etc.), but the whitelisting is (nearly certainly) done by Subsystem vendor. The Subsystem vendor is just the one who mounts the chip and surface-mount components to the PCB, sets the PROM strings, labels it, and sends it out the door -- the ones I have handy are Foxconn and Lenovo, [as I detail above](https://www.reddit.com/r/sysadmin/comments/1qmntsl/i_support_an_office_that_used_to_think_rebooting/o1uawx6/). WWAN cards I'd have to check if the firmware is loaded at init-time like WWAN, but it's the same either way.

Reply

[-]

Smith6612@reddit

The only reason I stated 6Ghz being a bigger mess than 5Ghz is because the GNSS requirements of base station equipment before certain 6Ghz channels can be used, and before higher power operation can be enabled. Some newer WiFi 7 Access Points I have will shut down the 6Ghz radios if they are unable to get a GNSS lock after a period of time, or if they can't speak to a server on the Internet to report the AP's operational location. Said APs also can't be used outside of the United States and you can't change the 802.11d region. A lot of 5Ghz equipment I've used in the past may have regional locks, but as long as you enable 802.11d you could use DFS without having to also use GNSS. You'd just get busted if someone noticed you running outside of regulatory rules and causing interference. DFS, I know there is additional certification devices need to go through to use that range. As for the DFS utilization problem, I have seen in past problems where cheap Android phones or the "hidden node" phenomenon triggers DFS Radar events. A lot of cheapo consumer devices (Smart TVs, older game consoles, IoT, many common streaming boxes) never got certified, and have DFS locked out in the firmware.

Reply

[-]

pdp10@reddit

Major laptop OEMs had/have firmware fifteen years ago that whitelisted only WLAN or WWAN cards that had their own-brand ID. I'm pretty certain the whitelisting uses the Subsystem Vendor ID and Subsystem Device ID, as seen in the `Subsystem:` line of an `lspci -vvv` or `lspci -q -vvv` command on Linux. My cards deployed on the test APs have Subsystem VID/DID of Foxconn 0xe0bb (QCNFA765), Foxconn 0xe0cb (MT7922), and Lenovo 0x5852 (RTL8852CE). Making things considerably worse, is that the system firmwares that do whitelisting, purposely give a vague error code at power-on when the unauthorized cards are present, instead of saying that they're refusing to boot because of the card. Again, information is hard to come by, and systematic information is unavailable. A third party would have to check dozens of cards with hundreds of laptops and firmware revisions.

Reply

[-]

pdp10@reddit

The major laptop OEM firmware blacklists are a different matter than the AX211 being a [proprietary CNVi card that won't work with non-Intel CPUs and chipsets](https://en.wikipedia.org/wiki/CNVi). An Intel AX210 or Intel BE200 can work without an Intel processor, but that doesn't mean that a laptop system firmware will allow a non-OEM WiFi card to work. This is determined by the laptop firmware, and has historically been ridiculously difficult to verify beforehand because it isn't documented by the OEMs. The only viable procedure is to swap and the card and see if it works; luckily the cards are inexpensive and units are commonly available if you're not picky about OEM. > sometimes 6Ghz won't work because of the ACPI tables missing information needed for Location Based Regulatory to work in the card's firmware. [Interesting; this is the first I've heard of this being an issue on Linux.](https://www.reddit.com/r/tuxedocomputers/comments/1mnph5m/ibp_14_gen10_why_6ghz_wifi_is_broken_and_how_to/) I've recently been unable to get RTL8852CE cards to emit on 6GHz and U-NII-4 and was going to revert to testing with Qualcomm cards on the same host platforms, given that no one seems to know what's going on.

Reply

[-]

Smith6612@reddit

Yep. The machines with CNVi (usually Intel vPro machines) are ones I'm aware of having fixed baseband support. Typically those cards are soldered down to the mainboard since you won't be swapping in another chip without swapping the entire CPU/Chipset. My laptop is an older AMD First Generation machine. The WiFi card is slotted and can be replaced. I just have to try another card as another poster indicated. I wasn't aware of the 6Ghz issue with Realtek cards. Only thought it was an Intel/Windows issue, as people would find that swapping Intel out for Realtek or Qualcomm would enable 6Ghz support to work in Windows. Linux? No idea... No experience with hardware that new on Linux yet.

Reply

[-]

nezroy@reddit

And you can get 2.5gbps on wired trivially cheap now, and 5/10gbps is absolutely within the normal business budget. Certainly not out of magnitude to the costs of a full WiFi 6E/7 setup. And the wired might actually sustain that for more than a single client. You're comparing the very latest wifi deployment to a 15 year old 1gbps wired deployment. And if your cabling was run in the last 15 years as cat 5e, it almost certainly supports a minimum of 2.5 easily and most likely 5/10 without issue too.

Reply

[-]

crysisnotaverted@reddit

In a similar vein, I have my homelab MicroPCs set up with 2.5g. I replaced the wireless cards with Realtek NICs for $15 a piece. Very easy. Honestly for a lot of installations, above gigabit doesn't make a ton of sense, it's cheaper and easier to run CAT6 drops and use cheap switchgear until it's totally necessary. 2.5 gig isn't a big enough increase to justify the upgrade cost, and most manufacturers went straight to 10 gig. 5 gig is a technological deadend as well.

Reply

[-]

pdp10@reddit

> most manufacturers went straight to 10 gig. The issue here is that compared to SFP+, 10GBASE-T consumes a lot of power, has significant distance limitations, and is inflexible. On the other hand, it's a familiar jack that doesn't scare amateurs, the flexibility situation improved with 10GBASE-T transceivers in SFP+ form-factor, and the format offers great possibilities for 2.5GBASE-T and 5GBASE-T support. This is what Apple picked for newer Macs. SFP+ is [amazingly flexible](https://www.youtube.com/watch?v=RFa7IWiqnb8) and has no distance or media limitations, at the cost of physical volume and no built-in compatibility with previous generations of UTP-based Ethernet. This is what enterprises use. Contrast with 2.5GBASE-T, which is just a straightforward upgrade to 1000BASE-T. No distance concerns, no power concerns, no figuring out how to connect the [Mac](https://support.apple.com/guide/mac-help/about-the-10-gigabit-ethernet-port-mchlcef4be50/mac) to the SFP+ enterprise switch at 10 Gbit/s. 2.5GBASE-T is pretty common on ATX motherboards and some SBCs, typically in the form of RTL8125B.

Reply

[-]

Icy_Conference9095@reddit

In a business case and for most home cases, does speed actually matter in a noticeable way? My previous work was in a medium post secondary institution. We had just finished upgrading to a 10gig pipe up from 4, we really don't notice a tying different aside from a percentage of bandwidth used across campus. Sure there are business cases for mukti-gig bandwidths for file transfers between sites, and I'm not saying it's not important in certain cases... But I have a few site locations in my new job that are still running on 100meg plans, one of which has 30-40 staff and we rarely hear any complaints. Before anyone says anything, of course I'm planning on upgrading a 100meg pipe; but sometimes they fight us on cost.

Reply

[-]

crysisnotaverted@reddit

Beyond me being a weird datahoarder with my house wired for 2.5gig, no, it doesn't make any sense. The funnier thing was, the PoE switch I used for that AP was only gigabit, so that greater bandwidth was only useful across the AP's wireless forwarding plane.

Reply

[-]

Ancient-Bat1755@reddit

10 feet from the router

Reply

[-]

SAugsburger@reddit

Good point. Some newer Wi-Fi standards can in some situations push beyond the 1G Ethernet port and that assumes that the device even has an Ethernet port in some cases. That being said I have seen an increasing number of workstations including at least 2.5G ports if not 10G although they're still mostly on higher end machines than most people would deploy outside niche power users. The cost for 2.5G ports on access switches has gotten a lot more reasonable that one switch refresh project they had rolled out 2.5G ports as they're likely not refreshing these switches for 7+ years.

Reply

[-]

itguy9013@reddit

Wireless for convenience. Wired for reliability.

Reply

[-]

pdp10@reddit

"Wireless for mobility." Every user knows they deserve convenience. Convenience is having your fixed devices always work, no matter whether the office next-door is de-authing your clients, or hosting a sales convention with 250 chatty devices.

Reply

[-]

Senkyou@reddit

> reliability And performance

Reply

[-]

iceph03nix@reddit

I've actually worked in offices where that was kinda true because they had an ancient switch, but a fairly new wireless router.

Reply

[-]

Lotronex@reddit

Yep, worked at an MSP, we'd find old hubs hidden in ceilings and behind desks all the time.

Reply

[-]

henk717@reddit

You did measure that to debunk it right? Because it can happen for sure. In my own house one cable got a bit loose and suddenly the wired connection was speed capped. Needed daily restarts to get the speed back and then at random times it would throttle again. Took me a moment to realize where it was happening but with the cable better secured it solved it. Same thing could happen with a faulty switch, incorrect qos, etc.

Reply

[-]

doofusdog@reddit

Yes... turned out that a long time back it may have been true, but this now was just a myth perpetuated in this branch. He didn't want to see the numbers. He was also adamant he wanted the old wifi ssid back, this new one just wouldn't do. there was a VLAN trunk misconfigured which confused us for a bit. The wifi and cable terminating on the same switch. Some months have now passed and all is well.

Reply

[-]

theWindowsWillyWonka@reddit

User's calling in requesting blue Ethernet cables. Idk how the rumor started but it seemed to be mythologized by every department that blue cables gave faster connection.

Reply

[-]

simAlity@reddit (OP)

I've come cross that as well. It's like if it's not blue.It's not the real thing. It's just sparkling internet?

Reply

[-]

ArrakisCoffeeShop@reddit

It's because they may have noticed that blue is used in server rooms, so it must be fast!

Reply

[-]

MyClevrUsername@reddit

That is just dumb. Everyone knows that red is the fastest color.

Reply

[-]

Sure-Squirrel8384@reddit

Red bypasses the firewall, of course it is faster!

Reply

[-]

PlayingDoomOnAGPS@reddit

Never seen Barry Allen wearing blue, have you?

Reply

[-]

ManiacClown@reddit

He's been a temporary Blue Lantern, so yes.

Reply

[-]

PlayingDoomOnAGPS@reddit

[\[image\]](https://i.imgur.com/TDPVQli.png)

Reply

[-]

Dhaism@reddit

thats why we use red cables for our wan connects! We tell users that the blue ones are fastest so we can push all the cheap slow blue cables to the end users.

Reply

[-]

jfoust2@reddit

And then there's October, when the pumpkin spice cables appear...

Reply

[-]

WorthPlease@reddit

Holy shit, Orks would be the best Sysadmins ever.

Reply

[-]

8BFF4fpThY@reddit

Encrypted connections are purple!

Reply

[-]

Serenity_557@reddit

One ork fought a robot and accidentally triggered a system reboot with a crowbar he found, upon rebooting it was corrupted, saw him as the "admin" and when he said "thems the enemy!" And pointed at the opposing army, the robot threw itself against its own forces. The crowbar is now known as the "computer stick" and you can smack any computer with it to make it do what you want it to.

Reply

[-]

TheThiefMaster@reddit

In our office red was for remote power control connections (think IPMI). You wouldn't want one for network! That said the red ones do go fastest.

Reply

[-]

OfficialDeathScythe@reddit

Excuse me. Yellow is actually the fastest which is why it’s usually included as the WAN link for home routers

Reply

[-]

iamalicecarroll@reddit

i thought yellow was

Reply

[-]

Serenity_557@reddit

Red is fast, but blue is lucky so.. Less packet loss, less errors while using the internet, that's probably why they thought it was faster lol

Reply

[-]

Bright_Arm8782@reddit

Coz Red Wunz Go Fasta, innit!

Reply

[-]

im-just-evan@reddit

Red is for spicy internet, orange and yellow are for the spiciest of internets!

Reply

[-]

DazzlingRutabega@reddit

Had a similar thing where religiously, every phone was connected with a black cable and a re-computer dock was connected with a blue cable. Users thought that they had to be connected with those specific colors.

Reply

[-]

lungbong@reddit

Like blue LEDs, they are the best LEDs.

Reply

[-]

Pazuuuzu@reddit

I wish companies would stop putting the brightest ever blue leds I've seen as a standby led in products...

Reply

[-]

mini4x@reddit

My PC monitors power lights BLINK bright blue when they are asleep, talk about the most annoying this ever.

Reply

[-]

Nesman64@reddit

The LEDs on my fridge make for an effective nightlight in my bedroom, 50 feet away.

Reply

[-]

lungbong@reddit

My broadband router has a very bright blue LED when it's connected, thankfully there's an option to turn the LED off, this should be spread wider.

Reply

[-]

Loading_M_@reddit

Tbf, there used to be a reason for this. Blue LEDs used to be much more expensive, so products using blue LEDs were often better than their competition (because they weren't cost cutting everything). Then someone came up with a way to make cheap blue LEDs, so all the dirt cheap products immediately switched to make themselves look higher quality.

Reply

[-]

NotMedicine420@reddit

Yh, 20 years ago. Not anymore.

Reply

[-]

TheThiefMaster@reddit

There were also later to market - red/yellow/green came first, and blue got a _second_ Nobel prize for LEDs because it had to be invented separately. So blue LEDs looked "new" for a bit too.

Reply

[-]

exhausted_redditor@reddit

I wouldn't be surprised if that office had repurposed existing (probably beige) Cat-3 cable runs for Ethernet, then started new Cat-5 runs with blue cables. After enough runs, the Cat-3 cables were probably only to be used for IP phones and similar, while the Cat-5 cables should be used for computers. Through enough tickets of users mixing the cables up, it became known that the blue cable was faster than the beige

Reply

[-]

geusebio@reddit

You could make a religion out of this ... no don't

Reply

[-]

Drywesi@reddit

too late, the Adeptus Mechanicus already did

Reply

[-]

NemGoesGlobal@reddit

The Orange Circle of Categorie 8? We will make network security our religion.

Reply

[-]

NemGoesGlobal@reddit

I don't know but in Europe most common cat cables for wall installations are happy orange no matter what categorie. Patch cables can have any color.

Reply

[-]

TheJesusGuy@reddit

Ours are cheap bottom tier grey 5e and all custom terminated

Reply

[-]

bentbrewer@reddit

The drops are crimped? I’m confused, are your patch cables all manually crimped?

Reply

[-]

TheJesusGuy@reddit

Sorry I mean it crimps them into the connector but does not cut them at the end so I have to bring some wire cutters and clip them off manually afterwards which means I cant get them as tight neat if I dont want a tonne hanging out the end. Now do this in the dark surrounded by spiders and you cant stand up.

Reply

[-]

NemGoesGlobal@reddit

Don't know about Britain but you can crimp your Patchcables by yourself if you want to. But you don't have to most of the time it's the better solution to buy factored patch cables. Most sysadmins here never did something like patching. They know about the Technique but aren't able to do it because they barely practice.

Reply

[-]

bentbrewer@reddit

Sure you can but why? Most network/sys admins wouldn't know a stranded vs. solid wire nor would they know which one to use. Punching down solid to a keystone or patch panel, sure go ahead, but crimping just leads to problems.

Reply

[-]

pinkycatcher@reddit

In the US orange is *generally* used for fiber. Ethernet is usually a grey/tan/black/blue. Blue is pretty common. Old cat3 was almost universally that weird tan. So this definitely makes a lot of sense to get passed down from older people who don't know the technical. And it's not illogical, in many industries different colors mean different technical specs.

Reply

[-]

Loading_M_@reddit

The users likely never see the ones in the wall, just the patch cable. My bet is they just happened to buy blue patch cables.

Reply

[-]

technos@reddit

> Through enough tickets of users mixing the cables up, it became known that the blue cable was faster than the beige one. That's exactly what happened at a place I worked in the late nineties. The phone was run using grey cat-3 cables that came with them, PCs used blue cat-5 cables made in-house. People noticed that if they accidentally used the grey cable with their PC, the network stunk, and combined that with the separate observation that (damaged) phone cables were replaced with blue ones and the phone would work better to get the insanity that was people stealing blue cables from others. Worse yet was the stealing that happened when a vendor shipped them a whole pile of orange cat-5e cables with some equipment. They didn't get used with the equipment so desktop techs used to keep a few in their tool bag as emergency replacements.

Reply

[-]

dhardyuk@reddit

As most voip phones had (have) a two port switch that would put a cat3 cable as the upstream link for the PC no matter which port it was plugged into on the phone. So blue would be the same as beige …… making both blue and beige equally bad.

Reply

[-]

Gabelvampir@reddit

There's a big company here in Germany that used to be state owned where you have to use yellow Ethernet cables in most offices. That's because they wired their network sockets in a weird non-standard way, and the matching custom cables are all yellow. But most people working there don't know that, they only that it only works with a yellow cable.

Reply

[-]

bentbrewer@reddit

At one point you would need a crossover cable to connect a router to a router (or computer) and a straight through cable to connect a router to a switch or switch to computer. Those days are over with auto-sensing ports but it used to be a thing for everyone.

Reply

[-]

Otto-Korrect@reddit

It is resolved... MOSTLY. I still have a problem once in a while and notice that somebody used a crossover cable. I replace it, problem gone. We get crossover cables shipped with some of our gear, so then tend to float around. I do know that if I see a red cable now to check it out.

Reply

[-]

Otto-Korrect@reddit

We had a telephone vendor (before my time) that set up our network and IP phones. They didn't have patch cables on hand, so made their own out of some random 4 pair cable. Who knows what standard it was, but the things failed constantly and weren't even crimped on well. It finally got to the point where we yanked every single remaining one and started over. That felt good!

Reply

[-]

EmotionalVegetable48@reddit

Depending on how old school the users were, I commonly saw Cat3 were grey and Cat5 were blue.

Reply

[-]

stopismysafeword@reddit

Holy shit, we always use blue for data, my manager is very strict about it and I always assumed it was because the company logo is blue but now that I'm thinking about it more, I think my manager might be on board with this theory too.

Reply

[-]

Muffinshire@reddit

We have a rhyme for our ethernet colour scheme, which starts "Wi-Fi, Wi-Fi never to be seen, that's why we use the colour green". It finishes with "5 metre blue, that'll do".

Reply

[-]

pppjurac@reddit

That bloke probably self describes as "An avid audiophile" ?

Reply

[-]

CasualEveryday@reddit

I haven't heard that one, but we did have an office make a big deal about us setting things up wrong and none of the phones worked because we used the wrong cables. I guess they'd had blue for data and orange for phones previously and the office manager had gone around and pulled all the black cables and now all the phones were just sitting there unhooked. We spent hours trying to figure out what was going on before someone drove over there to figure things out.

Reply

[-]

BudTheGrey@reddit

That may actually have it's roots in myth and legend. When I first started back in the early days of 10baseT networking, you would often see organizational standardization of blue for 10MB ethernet cables, and grey (often flat satin, but not always) cables for analog phone cables. In many installation, all were terminated in RJ45. Since the grey cables were often CAT3 at best, trying to use one for ethernet, especially as the conversion to 100Mb started, would give a poor, slow connection.

Reply

[-]

fencepost_ajm@reddit

"Silver streak" cable

Reply

[-]

SGG@reddit

If they previously had old cat3 cables (which are normally grey) then that could have honestly been true.

Reply

[-]

Walbabyesser@reddit

Orcs prefer red for speed 😏

Reply

[-]

thedbp@reddit

But blue is lucky so he may be on to something

Reply

[-]

Namaha@reddit

Possibly they used to have some old cat5e cables that weren't blue, and eventually upgraded to cat6 that were blue?

Reply

[-]

Junior_Resource_608@reddit

![gif](giphy|ZaWOItK1fy8J8hXq2Z|downsized)

Reply

[-]

ArrakisCoffeeShop@reddit

Side note: TBH, I've been in IT for decades and while rebooting computers to address some issue is important and necessary, a lot of help desk folks use this as some panacea without doing some real work in troubleshooting and diagnosing *why* the computer needs constant reboots. Replying to a ticket "Just reboot" and then closing it, doesn't inspire a lot of confidence, is lazy, and doesn't allow you to grow and learn. The rest of your comment is crazy tho. Updates being blocked and no restarts for months is wild.

Reply

[-]

Peredat0r@reddit

I was a sysadmin for a few years, and even I think rebooting is bad luck... Especially if they run Windows. Sometimes it is suicide 😉😄

Reply

[-]

Least_Gain5147@reddit

A global engineering firm with many public-facing servers, who insisted on patching and rebooting every 3 months out of concern for uptime. Seriously. That, or a medical company that used user's laptops for domain controllers, and the users routinely took them home after work. No vpns either

Reply

[-]

cjorgensen@reddit

Running two microwaves at the same time. We had a department that had a sign above the microwaves that said "Do not run both microwaves at the same time or you will blow a breaker." About once a week they would blow a breaker and the entire department would go black. 20-30 endpoints just shut down hard. Any unsaved data gone. All so someone could have leftover pizza. I told them, "This keeps happening you're going to lose a computer." Eventually this is exactly what happened. Power goes out, breaker reset, 29 out of 30 computers come back up. So now we have to hustle to find a computer to get on this guy's desk so he can work. I was tasked with making everything work. So I said, "This happens again I'm throwing those microwaves in the garbage." Everyone laughed. So when I delivered the computer I took one of the microwaves with me. Fuck it, I wasn't going to wait around for them to do it again, and you knew they would, since they did it over and over again. I dumpstered the microwave and the problem was fixed.

Reply

[-]

WorthPlease@reddit

Oh, I have a good microwave related story. I was a lone sysadmin in an office of about 100, plus my help desk guy. We rented out office space, so if we had random facilities issues we had to put in a request and it took days for them to show up, look at stuff, go huh, we'll get back to you. And then days later I'd have to explain to the other person the same thing. So I said fuck it and just started fixing everything myself. Security doors, cameras, TV's, etc. One day the office manager/front desk assistant came up to me and said there was a problem with the microwave in the meeting room kitchen. It said "FOOD" and she couldn't microwave anything. Google what the problem is when a GE microwave says "FOOD" and won't let you start it. >!It means the door isn't fully latched closed!<

Reply

[-]

fencepost_ajm@reddit

"Sorry, the microwave doesn't think your meal is up to its standards. You're going to have to eat it cold."

Reply

[-]

WorthPlease@reddit

I agree that is a silly message, why didn't they program it to say OPEN.

Reply

[-]

Sure-Squirrel8384@reddit

Or how about "DOOR"?

Reply

[-]

WorthPlease@reddit

Hey GE, hire us.

Reply

[-]

Walbabyesser@reddit

What kind of shitty electric wiring did they have there?

Reply

[-]

cjorgensen@reddit

The building was over 100 years old: https://en.wikipedia.org/wiki/Register_and_Tribune_Building

Reply

[-]

letsgoiowa@reddit

Huh, real neat to see a piece of local history in a random thread here! I'd believe it lol

Reply

[-]

cjorgensen@reddit

I worked in that building for 13 years. I set the sprinklers off once and flooded the third floor. Good times.

Reply

[-]

BatemansChainsaw@reddit

if you're the guy that did that in '93 my old colleague would like a word with you :P

Reply

[-]

cjorgensen@reddit

Nah, I did it sometime in the 2000's. I had five of six Mac G4s in boxes on a cart. They were taller than I was. I was going down a sloping hallway and the ceiling came down toward the end. The boxes hit the sprinklers and 100 year old oily sticky brackish water blew all over me. It was disgusting. I got the computers out of the water and called my boss. It sucked.

Reply

[-]

StaticFanatic3@reddit

Two microwaves will trip any standard circuit in the US. Doesn’t matter if the wiring was put in yesterday

Reply

[-]

Walbabyesser@reddit

This is… awkward. I use a microwave at work daily and there is another one right next to it. At lunch they both run parallel. BUT I‘m not in the US… 😏

Reply

[-]

dustojnikhummer@reddit

Two microwaves (if they are weaker, like 1200W) can be fine on a 12/16A 220-240V circuit. US circuits are only 110-120V.

Reply

[-]

StaticFanatic3@reddit

To be clear we do have plenty of kitchens and company break rooms with more than one microwave. It’s not like it’s rare to have 40 amp circuits put in place for that very reason. Or even 240volt circuits like yours 😉 But it has to be planned for and if someone just brings in a second microwave and plugs it in to a 20amp or, god forbid, 15amp circuit then yeah it’s gonna pop

Reply

[-]

Walbabyesser@reddit

Agreed

Reply

[-]

dustojnikhummer@reddit

Old building with aluminum wiring and 6/10A breakers I suppose?

Reply

[-]

dustojnikhummer@reddit

Trip or literally blow up a breaker?

Reply

[-]

cjorgensen@reddit

Trip. We would just reset the breaker. IT knew where every one was in the building as this wasn't uncommon. Space heaters and morons plugging in powerstrips into powerstrips.

Reply

[-]

pppjurac@reddit

Remove one of mw and be done or a a job for electrician. A simple breaker on the wall that connects both mw and trips when current goes over 20A .

Reply

[-]

alucardcanidae@reddit

We had the same issue, but our electricians separated our 4 microwaves to different breakers.

Reply

[-]

Phayzon@reddit

What even was the point of having 2 microwaves if only one could be ran at a time?

Reply

[-]

GMginger@reddit

RAIW1 - Redundant Array of independant Microwaves.

Reply

[-]

thatpaulbloke@reddit

So that you can seal one shut after someone microwaves fish in it.

Reply

[-]

Krigen89@reddit

Empty/fill one up while the other one runs? Also, they probably didn't know this would happen when they bought the 2nd one.

Reply

[-]

sparkyblaster@reddit

Yeah removing one microwave seems like it solves a lot of problems. Why even keep two? Mess spread across two units instead of 1?

Reply

[-]

SikhGamer@reddit

I would have taken the second microwave as soon as they blew the breaker a second time.

Reply

[-]

cjorgensen@reddit

I wasn't management. The early solution was to put up the sign, then when it kept happening I' guessing more signage, but I don't remember. I never understood why they needed two microwaves if they couldn't be run at the same time to begin with.

Reply

[-]

JaschaE@reddit

An Admin I know helped plan out the layout and infrastructure of our hackspace. The electronics area (4soldering workspaces) has it's own circuit breaker box, each space secured separately, each place with grounding mats. He was disappointed to learn "But it's cozy here" supersedet all this setUp so somebody dragged the soldering station into the lobby, shorted their project and killed the lobby lights and several parts of the network equipment.

Reply

[-]

ck17350@reddit

An entire tech support department in the 90’s telling customers to defrag their hard drives because that’s what was keeping their modems from connecting to our modems. I had to rewrite nearly all of their troubleshooting documentation because it was pretty much all make believe nonsense.

Reply

[-]

simAlity@reddit (OP)

How much pushback did you get over this?

Reply

[-]

ck17350@reddit

Next to none, considering that since they had no idea why they were telling customers this in the first place, being told to do something different wasn’t really challenging anyone’s beliefs. They were poorly trained unfortunately. When I was able to provide steps that actually kept the customers from calling back in an hour, our techs happily adopted this new information. Granted, I also took the time to explain why they were doing these new things to help them understand and become better techs. Many of them went on to be quite successful in the IT field. It was a bunch of teenagers getting their first jobs or paying for college so they had no background before coming on board. Good bunch of people, honestly.

Reply

[-]

Scared_Bell3366@reddit

I was told the defrag trick was to occupy the customer’s time while fixing something on the back end.

Reply

[-]

ck17350@reddit

Yeah, I think that was the original intent, keeping the customer busy for a while and hoping the issue would fix itself. But with the turnover, defrag and other strange things became part of the mythology of how computers must work.

Reply

[-]

simAlity@reddit (OP)

That's pretty good. I was asking about management though. Any problems there?

Reply

[-]

ck17350@reddit

Whoops, I misunderstood. Management had no issues and were happy to see things getting revamped by someone who cared enough to do it.

Reply

[-]

aprimeproblem@reddit

Did they feed that to Copilot? Would explain a lot!

Reply

[-]

Poohbutt2005@reddit

Leave well enough alone. It's working now. Let's not mess with it. No, that clicking sound you hear is the hard drive dying in the PC that runs your door access, which ties into the time clock which ties into payroll. The hard drive died. The database was not backed up.

Reply

[-]

BatemansChainsaw@reddit

sometimes letting things fail and not backing them up teaches them a lesson and gets their attention for the next time.

Reply

[-]

hotfistdotcom@reddit

Sometimes you can make a private backup, then tell them they really needed to back up, but maybe you can recover it if you disassemble the whole drive and try really, really hard. You'll have to use your lab at home for it and it'll take a week or so of very hard hands on work, but you might be able to pull it off. And then, you know, do what you want for a week, take some pics of platters out of a drive and maybe a drive with a bunch of alligator clips on it and then at the end of the week restore your private backup.

Reply

[-]

Poohbutt2005@reddit

Yeah, that's what I would have liked to do. I was new and has explicit instructions saying "let's just leave it for now and see what happens." I do not miss those days.

Reply

[-]

Poohbutt2005@reddit

It was my first IT job back in the day. I knew enough to CYA. I sent an email to my manager and HR. My manager agreed to the replacement but was overruled by HR. It was a strange hierarchy at this company. The hard drive failed. Fingers were pointed. Emails were referenced. Money was spent.

Reply

[-]

BatemansChainsaw@reddit

> manage [...] overruled by HR what in the world? damn...

Reply

[-]

roboto404@reddit

The only weird behavior i’ve been dealing with is deciding that the office should be open concept style, then a month later, complain about people being too loud and not being able to concentrate.

Reply

[-]

geusebio@reddit

I had to talk to my therapist about how Office Space's idea of dedicated workspace and divider walls, while back then was a dystopic nightmare, is actually currently high-fantasy and aspirational. Imagine having that level of job security

Reply

[-]

dustojnikhummer@reddit

Oh hey Bob!, Bob.

Reply

[-]

geusebio@reddit

I used the bobs asking some jackass "fractional CTO" salesman the other day. Why would I pay someone to be fractional CTO to reply to shit with chatgpt.. I can make a script that does that.

Reply

[-]

anothercopy@reddit

I worked on older servers that didnt get a lot of reboots. Think Sun, Aix etc that sort of thing. Also a lot of stuff could be done online without a reboot and a lot of was done. With a big team managing these servers you cant really know what was changed between the last reboot. Some things were admin fuckups, some were vendor fuckups. So rebooting or especially rebooting after patching could give you lots of surprises. Also on these older servers you would find things failing on POST or a drive failing in the mirror. The worst was that you didnt see before your reboot that the mirror was already degraded and the other barely hanging drive died during the reboot. Then you have an allnighter usually to replace and restore from backup. So yeah rebooting not always fun.

Reply

[-]

Sure-Squirrel8384@reddit

Root cause is lack of hardware monitoring.

Reply

[-]

anothercopy@reddit

I think BMC Patrol was rolled out back then. The problem is that while it covered all platforms, never really in depth. Like it didnt know how to query all the specific disk mirroring software. Bear in mind only on Solaris you had 3 different software versions of making disk raids and at least 2 different ways to do it on hardware (disk controller) level. And a lot of those disk controllers for quite some time didnt have a way to report back to the OS so yeah. Anyway my custodian times are gone ;)

Reply

[-]

Sure-Squirrel8384@reddit

Yeah, we just wrote custom scripts for each OS/platform type if there wasn't native monitoring tools. Then rsh or whatever to execute every X minutes and record the output which we standardized to make it look like what could be ingested natively..

Reply

[-]

NoClownsOnMyStation@reddit

Stop writing everything on paper. We would be in meetings and people would bring entire notebooks to reference stuff. I’m a huge advocate of creating an engineering notebook as you work but if you’re bringing a physical notebook to reference data from there’s probably a faster way to do it.

Reply

[-]

Binky390@reddit

Not an office wide problem, but I used to work at a university and a high level dean had his assistant call me to complain that emails he was sending to trash were being permanently deleted in Gmail. I was like well yeah, it’s trash? I can’t stop Google’s behavior. She was like then what is he supposed to do when he needs an important email? I was like don’t trash it? He was using the trash as a folder. So odd.

Reply

[-]

sarat023@reddit

Kind of the opposite of this, remember when in MacOS you dragged a disk to the trash to "Eject"? My dad was horrified when he saw me do that and was sure I'd destroy the disk that way, had to slowly explain that's an intended feature. Then I really messed with him by dragging the main disk to the trash, he seemed to think the computer would self-destruct the moment I let go of the button.

Reply

[-]

Sure-Squirrel8384@reddit

I've heard that many times... that the trash was somehow a short-term storage location. SMH, really? Would you file your real papers in a trash can and get made when it was emptied each night?

Reply

[-]

Otto-Korrect@reddit

OMG, we had an older woman who was suddenly crying that all of her emails were gone. Turns out she was using her trash folder as long term storage, and we'd purged trash folders to free up some storage space (some users had 10s of GB and had never emptied the trash). I STILL don't think she ever understood what that wasn't a good idea. Thankfully, she retired soon after.

Reply

[-]

NDaveT@reddit

> She was like then what is he supposed to do when he needs an important email? The correct response is to teach them how to create folders. You'd think it would be intuitive but it's not. It's only intuitive if you have basic computer literacy.

Reply

[-]

Binky390@reddit

We did. This was a year into the Google migration. But teaching them folders wasn’t even the issue. The trash isn’t a folder. Ever.

Reply

[-]

stopismysafeword@reddit

I've came across this with the Windows recycle bin too, I had emptied it because their laptop was running out of storage and they went nuts!

Reply

[-]

Ballbag94@reddit

I've had a lot of users doing this, back in the days when PSTs were stored locally and caused issues with Outlook when they were too large I had to shrink a user's file. They had hundreds of deleted emails so I went to delete them but on my way to do that I decided to ask for some reason, turned out they were using it as an archive location and had critical info in there, they didn't listen when I suggested they not do that going forwards

Reply

[-]

TU4AR@reddit

This is a rite of passage for anyone in IT. Same with the classic : I swear I turn my computer off with the uptime of 6 months.

Reply

[-]

stas-prze@reddit

I mean nowadays with Windows Fast Startup they might not even be wrong, I think that only does a proper clean reboot in the case of a restart, not shutdown.

Reply

[-]

hotfistdotcom@reddit

It's crazy how often I've seen people doing this. And then you suggest something like "lets make a folder called pre-trash or archive or something" but no they want the trashcan. I feel like it has to be some kind of weaponized incompetence thing

Reply

[-]

jleahul@reddit

"I put my sandwich in the garbage because I wanted to save it for later, but now it's gone!" 🤦‍♂️

Reply

[-]

dalgeek@reddit

I had a user do this with Exchange. They put everything into the trash to keep their inbox clean then search for them later if needed. Their mailbox was over 1GB in size and 90% of that was in the trash.

Reply

[-]

SikhGamer@reddit

This is crazy common.

Reply

[-]

zaphod777@reddit

I mean, Google's use of labels instead of folders isn't very user friendly or intuitive.

Reply

[-]

Shanix@reddit

> He was using the trash as a folder. So odd. Isn't this an _old_ holdover? I vaguely recall people using trash/recycling as another folder because it wasn't put against your storage quota (or wouldn't be auto-deleted, one of the two).

Reply

[-]

HairyMechanic@reddit

It was rife when I started in the tech world a decade ago and I used to get various reasons for it - storage space was always the main one. A few people had it as a force of habit which eventually caught them out when deleted/trash locations would be deleted. Never happened twice for any person, mind you, so a vital learning curve.

Reply

[-]

PlayingDoomOnAGPS@reddit

I've seen this dozens and dozens of times, mainly in Outlook. We'd have these massive PSTs getting corrupted (because that's what PSTs do) go check the Trash and they're using it as an archive. I've seen more offensive stupid behavior but this is one of the most baffling stupid behaviors for me.

Reply

[-]

bostonronin@reddit

This and using the Drafts folder for the same purpose is disturbingly common among users in certain environments. Even when they know how to create new folders. Had to deal with this when switching users between Outlook and Gmail, as drafts and trash don't usually migrate.

Reply

[-]

MortadellaKing@reddit

Years ago I had to restore an exchange database because that company had retention polices set to empty everyone's deleted items after X amount of days. Well turns out this small subset of users were using deleted items as their storage. Because, you guessed it, previously we were on lotus notes/domino and that didn't count towards their quota. I actually pressured management to let me not restore these mails, but some of these people were extremely squeaky wheels at that org.

Reply

[-]

dracotrapnet@reddit

I had someone complain all her friend's emails were gone. I asked where did you store them? In the trash. I asked if the trash doesn't get picked up in her office every other day cuz it does in mine. Yea, it's gone. Well you the email trash is special, it has a 30 day countdown before it vanishes and I can't do anything about it. Save your friend's emails in a personal folder. You know that says a lot about what she thinks about her friend's emails.

Reply

[-]

KStieers@reddit

Had a paralegal doing the same thing... and complaining about space in her mailbox. And someone used the actual paper recycle bin under their desk that way, and something got tossed so we all had to take our own recycling to the trash for a while. Like what the fuck!!!

Reply

[-]

j550@reddit

Had to take away their ability to make bad decisions. Removed the shutdown option with group policy because a surprising number of folks would shut off their remote PC and then complain the next morning of not being able to remote in.

Reply

[-]

Sure-Squirrel8384@reddit

Same, and then we had to create "Reboot" icons on their desktops because the GPO sucks and blocks Shutdown and Reboot (or used to).

Reply

[-]

the_gum@reddit

https://preview.redd.it/7ftd71iylnfg1.png?width=781&format=png&auto=webp&s=0eb031476bf48333562e34044fab999a32e4a1a6 The irony ;)

Reply

[-]

Sure-Squirrel8384@reddit

Root cause: pushing patches too soon and skipping test / QA.

Reply

[-]

BeepoZbuttbanger@reddit

The exception here would be old servers that are out of warranty. I mean, it’s your box and you’re free to do what you want, just don’t expect I can magically get it running right away when it doesn’t come back up.

Reply

[-]

Sure-Squirrel8384@reddit

Reboot, not shutdown. Powering off will reduce heat for a bit and cause problems. Plus you know those CMOS batteries are long dead.

Reply

[-]

bcredeur97@reddit

Regular reboots have always made things better in my experience. The only exception is when people (*cough* developers *cough*) setup something and then don’t plan on the computer rebooting — but that’s why I stress anytime something new is setup you need to test a reboot and be sure everything comes up correctly automatically

Reply

[-]

Sure-Squirrel8384@reddit

Before Systems takes ownership of a system from the Devs one of the tests is a reboot. Devs should have proper shutdown and startup scripts so that Systems can patch/reboot without having to do anything further.

Reply

[-]

Ssakaa@reddit

That's a good reason to reboot (and re-deploy containerized workloads) monthly at a minimum, on top of just *actually* making sure all installed patches take effect. Configurations that've deviated from the reviewed, approved, changes that are reflected in code shall be cleansed with holy fire, and if they're not both in a test environment *and* reflected in a pull request in the appropriate repos, the dev responsible for them ought to be cleansed with holy fire too...

Reply

[-]

electricheat@reddit

one small office i took over, the secretary would walk over and power off the file server any time she had internet problems took a surprising amount of work to get her out of that habit their shitty dsl connection would regularly drop for about the time it takes to power cycle the server and go for a smoke, so she was convinced it worked

Reply

[-]

Sure-Squirrel8384@reddit

Locked cabinet? Hah, I'd never trust staff with direct access for just this reason. Next thing you know they're charging their cell phones in the server USB ports.

Reply

[-]

Sure-Squirrel8384@reddit

LOL, reminds me of the submarine guys who think it's bad luck to wash a coffee mug. From a sysadmin level when I started they refused to change the default domain administrator account password. "We tried to change it once and it broke everything". Uh huh. Needless to say, this was a priority over the next year to identify and change all services that were using this account to their own dedicated and documented service accounts; plus annual password rotation of all accounts The default domain administrator account is now disabled as well. We even found that the password had leaked to the help desk and they used it routinely (SMH).

Reply

[-]

TTLogs@reddit

The weirdest thing I’ve ever seen like this was a small company where trackballs were mandatory. New starter? You gotta learn to use a trackball. No mice allowed. They had gotten into their heads that trackballs prevent hand strains / injuries. It was the most ridiculous thing I’d ever seen. BUT, once I’d been forced to use a trackball I’ve actually never gone back

Reply

[-]

andytagonist@reddit

*”We’re attorneys, we’re smarter than you.”*

Reply

[-]

suicideking72@reddit

I had a client disable automatic updates 'because they'll crash your PC'. He was one of the managers, so half the people at that location had updates disabled. Then they would call you with weird app problems and wouldn't let you install updates. That was pretty annoying. Even after explaining why updates are important, he would just nod, then not let you install updates.

Reply

[-]

Otto-Korrect@reddit

Computers not so much, but I still get anxious whenever rebooting a physical server, especially one with a long uptime. That gives it time to accumulate bugs with updates or hardware issues, bot of which can lead to devices failing. I've see the 'boot device not found' message more than once on a production server. Of the system looking for a PXE boot source after everything local fails. We once had an AS/400 with an enormous uptime. When we had to shut it down to replace a RAID interface, the tech warned us that after being up for that long, we may lose a few drives. Sure enough, we lost 3 drives and lost the array. Had to do a full system restore then a restore of our most recent data. OH, and our system restore was on a tape format that we didn't have the drive for anymore, so we had to scavenge up the right drive so we could use that tape. I think I was there until well after dawn that next morning, but got it back online just before the start of business. So yeah, now I hold my breath every time I do 'restart' on a server.

Reply

[-]

Upbeat_Whole_6477@reddit

Had to deal with an IT manager and Sys Admin who did not patch servers or workstations because “the updates corrupted things…”

Reply

[-]

tropicbrownthunder@reddit

Can't really blame them. Back in my time patches were more mayhem then helpfulness. (I'm looking at you SP2 for NT4) And well Windows 10 and 11 have really kept the tradition

Reply

[-]

TheCollegeIntern@reddit

Can’t say I blame them. It’s an old school way of thinking. I don’t agree with it bit if your starts screaming Notepad isn’t working you might be unreasonable if you work in an unreasonable environment.

Reply

[-]

Walbabyesser@reddit

Total normal behavior 🤡

Reply

[-]

fencepost_ajm@reddit

Eh, Microsoft is trying to return us to that, see the January 2026 update that broke little things like Outlook with PST files in Onedrive Documents (the default for the past several versions). Ironically only affected my most important customer, SWMBO who uses it for a personal domain.

Reply

[-]

Ad-1316@reddit

if there's a problem, look for the yellow cable. It was the crossover cable.

Reply

[-]

FutureGoatGuy@reddit

When I came into this org, the (at the time) head of IT (and all previous heads) did not use GPOs because "it could brick the machines". Deployed printers? No. Groups membership that mapped network drives? Nope. Base GPOs? Yes, and thats it. So after that head left I spent a week creating base level GPOs so that I could stop having to do things like map printers\\drives manually for users but it was a long two years of waiting.

Reply

[-]

simAlity@reddit (OP)

I'm gonna DM you....

Reply

[-]

hellofairygodmotha@reddit

It would cost too much money in downtime to have user’s change their passwords lol

Reply

[-]

simAlity@reddit (OP)

Doctor's office?

Reply

[-]

hellofairygodmotha@reddit

Engineering 🤯

Reply

[-]

GnawingPossum@reddit

I've seen an MSP that did not believe in domain joining workstations, everything in workgroup mode, mismatched passwords between local computers and AD, etc. I've seen another MSP sell Netscreen and Fortigate firewalls without security service subscriptions. I've seen so many businesses refusing commercial AV "because Windows comes with a free antivirus". "We don't need fancy backup software, just clone our server's drive once a month..." And so many more I wish I could just forget over the past decades. And yes, in all of these cases, stakeholders were informed of the risks and offered even cheaper alternatives and still insisted on going forward with their bad practices.

Reply

[-]

gib_me_gold@reddit

\>I've seen so many businesses refusing commercial AV "because Windows comes with a free antivirus". More than enough nowadays, though.

Reply

[-]

GnawingPossum@reddit

For personal use, sure. But the lack of visibility and management in a business context is not great. Sure you can lock it down with GPOs and a SIEM to replace the missing central reporting, but realistically, these same types of businesses are unlikely to have a SIEM.

Reply

[-]

highdiver_2000@reddit

Don't laugh. Harddisk engineers never shutdown their cimputers

Reply

[-]

AccurateElk4233@reddit

Weirdest one for me was a place that believed logging out breaks software licenses, so everyone just locked their PCs for months. Uptime was measured in weeks, updates disabled, apps half-broken. Same fix as you: force updates + scheduled reboots. Magically, all the “bad luck” disappeared once machines were actually maintained.

Reply

[-]

graywolfman@reddit

I had one CIO that thought we should shut down and power back on all networking equipment right before the holidays. Same CIO thought we should rename all servers and networking gear so you couldn't tell what the role or model was by its name. He thought that naming something like 01DC01 meant a hacker being in our network would more easily know what to target. Like DNS, DHCP packets, or general network traffic wouldn't tell them, anyway, if they're that far in. Have a company telling our director we shouldn't put humidity control in an MDF that's getting built because they're going to put antistatic coating on the floor. /sigh

Reply

[-]

Darkhexical@reddit

I mean I've heard most attacks happen during the holidays so makes sense when you think about it.

Reply

[-]

wazza_the_rockdog@reddit

> Holiday downtime is actually the perfect time for a reboot; fix the bugs when nobody is in the office to complain about the outage. As long as your gear is under warranty and the vendor is working. Not a shutdown caused issue, but previous employer only had a single split system air con in the server room, it seemed to know when a public holiday or annual shutdown period was coming because it always seemed to have faults just beforehand. Calling in aircon companies to fix it on the last day before every aircon company are off for christmas or whatever was always painful (and I'm in Aus, so Christmas time is summer for us, and gets bloody hot - so keeping them running without AC would not work out well for us). Same company was anti-cloud and wanted 24x7x365 uptime, but definitely didn't want to pay for anything even approaching that.

Reply

[-]

graywolfman@reddit

>Holiday downtime is actually the perfect time for a reboot; This was retail, I should have included that little detail lol. Bad time. Very bad time. Also, enterprise routers aren't really made to 'reboot' unless you are responding to an issue or doing firmware upgrades, etc.

Reply

[-]

Darkhexical@reddit

Not made to reboot? Just because it takes a while to come back up doesn't mean you shouldn't reboot it. Ive had reboots fix quite a few things on enterprise routers, switches, and firewalls.

Reply

[-]

SAugsburger@reddit

Cisco over the years has had a number of bugs that a reboot would at least address the symptoms (e.g. I remember an older IOS XE bug that had a bug that would eventually drive CPU to 90%+ even without doing much to justify, another bug where the management plane became completely unusable after enough uptime), but the reboot without a software update that patched the issue wouldn't really fix the problem. It just kicked down the road until you hit the bug(s).

Reply

[-]

Ssakaa@reddit

Yeah... I've definitely seen some slow-growing memory leak like misbehavior from them. A monthly scheduled reboot did *wonders*.

Reply

[-]

graywolfman@reddit

Yes, like I said, for fixes and firmware. Generally, you don't need to reboot if they're all running perfectly, though ...

Reply

[-]

Angelworks42@reddit

We had some decades ago who obfuscated database names for this reason - to this day the oprd database is production.

Reply

[-]

flummox1234@reddit

I mean bufferbloat is an actual thing so it's not that crazy. The naming was pretty common back in the day but is a bit outdated given tools that have evolved now so maybe it's a hold over from a previous IT era.

Reply

[-]

Ssakaa@reddit

Security by obscurity was never a real benefit. The *closest* thing to value from naming being "accurate" or not is making a honeypot grab attention more easily.

Reply

[-]

Rocky_Mountain_Way@reddit

About ten years ago, my employer had an administrative assistant put everybody’s keyboard into a dishwasher on rinse mode to clean them. Some kinda worked after, but most had to be e-cycled

Reply

[-]

pdp10@reddit

Twenty years ago, we ran a test to see if that worked. * All keyboards came out super clean. The beige models showed the most dramatic difference. Therefore, we were very eager for this technique to be viable. * Most or all rubber-dome keyboards had at least one key that didn't work properly afterwards. At the time, we speculated that residue from the dishwashing powder might be causing this, and wondered if no-detergent cycle would work perfectly. Teardowns didn't shed any real light on the matter, but we put a pause on the project before going all the way to electrical troubleshooting. * My recollection is that all of the keyboards tested were PS/2 keyboards.

Reply

[-]

pppjurac@reddit

> residue from the dishwashing powder Probably that and minerals from water. You can was electronics, but it has to be distilled and demineralized water with special detergent.

Reply

[-]

Rocky_Mountain_Way@reddit

Interesting data. Thanks.

Reply

[-]

Krigen89@reddit

Back in the 90s my father's friend would disassemble those 2.4ghz wireless phones, remove batteries and throw the PCBs in the dishwasher. As long as you made sure they were dried properly before putting the battery back, it actually improved call clarity somehow. Lol

Reply

[-]

thedanyes@reddit

It's a great idea in concept, and could probably reduce the spread of disease. They just need the right hardware to support it.

Reply

[-]

Walbabyesser@reddit

It works, but it helps to remove the electronic parts inside before that 😆

Reply

[-]

simAlity@reddit (OP)

I've actually heard of that trick being effective. It may be a myth or something left over from the age when things were made to last.

Reply

[-]

Rocky_Mountain_Way@reddit

Oh, yeah the old IBM keyboards with the big DIN connectors (before the PS/2 and USB connectors) WERE built to take abuse like dishwashers, but the new cheap, light, USB keyboards break really easily

Reply

[-]

boli99@reddit

>bad behavior you have had to cure on an office wide level? ... to break several groups out of the mindset that 'the only way to fix a problem was to physically visit the computer with the problem' It's not 1988 anymore.

Reply

[-]

PlayingDoomOnAGPS@reddit

Using CAPS lock instead of holding shift to get a single capital letter.

Reply

[-]

stas-prze@reddit

As a kid I had no clue you could select stuff with shift + arrows, so I would just type Dropbox links from one window to the other to download a file. I think that's worse lol.

Reply

[-]

PlayingDoomOnAGPS@reddit

Bro, I legit did not know you could select text with shift and the arrows. I use KB shortcuts to copy, paste, cut, and so on but selecting text has always been a mouse thing for me. Why would you retype them instead of using the mouse to select the text?

Reply

[-]

stas-prze@reddit

It's because I'm blind, I can't really use the mouse effectively. I know there's blind people who do try to incorporate the mouse in to their daily lives and succeed to some degree, but that pain and angwish is not worth it for me when most things I'd ever want to do are doable with a keyboard. I pretty much only keep a mouse around for emergency situations like having to hold the mouse button to open the tray on openfirmware macs (Why'd you do that Apple!) or random UI glitches which prevent a screenreader from reaching certain things, which is less and less of a problem as time goes by.

Reply

[-]

PlayingDoomOnAGPS@reddit

That's fair. I know for sure I'm going to be using that shift-arrow trick from now on! Shit, I wonder how long that's been a thing and I never knew...

Reply

[-]

thedanyes@reddit

I agree with them. Bad luck because it lowers your productivity. The best luck would be an operating system that didn't become increasingly insecure and unstable the longer it runs. Kernel live patching was a big thing 10 years ago. Microsoft still hasn't got the memo. https://www.youtube.com/watch?v=SYRlTISvjww

Reply

[-]

hotfistdotcom@reddit

stability, sure. But do you have a source for less secure?

Reply

[-]

thedanyes@reddit

My point is that you are required to reboot for security patches in Windows. Let me know if you still want a source for that.

Reply

[-]

hotfistdotcom@reddit

Ah, I thought the implication was that somehow higher uptime results in reduced security directly, rather than not patching is bad. Like it gets all unstable and somehow, malicious actors can smell it

Reply

[-]

hankhillnsfw@reddit

Honestly tho rebooting Linux can fuck your day up. Rebooting Windows is like troubleshooting step 1 lol

Reply

[-]

stas-prze@reddit

This brings me back to the days of being a teenager and having an obsession over Manjaro for some reason. Because I'm blind I couldn't just install it with a normal image since that didn't include sppeech, so I used Vagrant to deploy a VM which was over a year old from an image someone made, then pacman -Syu, generate iso and not reboot as to not make it break. If the image didn't work, go through all of this again untill I got a working one. To my astonishment some legend still keeps maintaining the talking images despite me pretty much not doing anything with them for over 5 years.

Reply

[-]

hotfistdotcom@reddit

Twice monthly restarts? It sounds like the superstition is alive and well brother

Reply

[-]

fencepost_ajm@reddit

Long ago, but an office with some users who'd decided that shutting down took too long so they just held the power button in. I think this may have been in the WfWG 3.11 days. They had something in place that was doing a text-mode GUI disk check at every startup.

Reply

[-]

stas-prze@reddit

That was me at 13 - 15. I used to use a Macbook Air and would often have Windows running in a VM as OS X as the host. For some reason I would W/O fail shut it down by holding the power button pretty much every single time. One such incident caused my VM to die. I think I only broke that habbit over the last 2 - 3 years or so really, I still remember doing it with my poor 2008 Mac Pro when that was my primary desktop in 2020 to 2022.

Reply

[-]

da_apz@reddit

When SSDs were well into the majority, I had a small business customer with maybe 10 workstations total, who insisted they don't want SSDs because they can't trust the chips will hold the data. Only the HDD technology was good enough. They got sold couple of laptops, which had then be swapped with a HDD. I was out the company before I learned how they reacted to mSATA and NVMe times.

Reply

[-]

ThePerfectLine@reddit

The one point in favor of that is that spinning disks fail slowly. They tend to start having problems and if you see them early enough you can usually save the data. SSD if going gang busters minute and then ☠️the next. At least has been my experience.

Reply

[-]

wrestler0609@reddit

I would just tell the users that Static Discharges over the weekend from the lower troposphere are causing interference. I always get a oh ya that must be it. To be clear that is lightning…

Reply

[-]

omnichad@reddit

Cosmic rays can cause bit flips in RAM. Cosmic rays were also found to be the trigger to actually spark a lighting strike during favorable conditions. Therefore your lie would actually possibly be correct correlation even if it's incorrect causation.

Reply

[-]

Weird_Lawfulness_298@reddit

Web guy decided that passwords had to be unique when logging into the e-commerce site. So, two unrelated people couldn't have the same password. In addition, it would tell them the password was already used and to try another.

Reply

[-]

omnichad@reddit

That's a clever way of preventing people from using insecure passwords. Not good, but definitely clever. If two people have the same idea it can't be that secure.

Reply

[-]

WinterFamiliar9199@reddit

I had the opposite. Everyone shutdown computers everyday to save power. Try doing updates for 3000 powered off workstations. Then when they come in the next day everyone is pissed at their forced reboot in the middle of the day.

Reply

[-]

omnichad@reddit

Sleep with a scheduled wakeup would have 99.9% of the same power.

Reply

[-]

z_agent@reddit

We want to go wireless cause then the building will weigh less....

Reply

[-]

simAlity@reddit (OP)

What....?

Reply

[-]

omnichad@reddit

The Internet doesn't weigh anything because it's wireless. Or something like that.

Reply

[-]

Scared_Bell3366@reddit

I worked on semiconductor equipment for many years. Customer figured out how to check the disk fragmentation on the computer running the equipment and wanted it defragged. It took several hours and the equipment was offline the entire time. The application running the equipment wrote many small files and would fragment the drive in about 10 minutes. Colossal waste of time and money.

Reply

[-]

Optimal-Archer3973@reddit

That CD drives were coffee cup holders and would also keep coffee hot on the managements computers..

Reply

[-]

fonetik@reddit

I just took a job late last year and left it soon afterwards when I learned that I was going to be the pilot for their new laptop model. Not the new model of laptop, the company's new business model that was laptop based. I was the first laptop-based employee in the org, and I'm the IT guy that will be expected to provide senior support. A decent sized bank with over 1000 employees. Everyone was still on towers or thin clients. No remote access for employees until COVID, then they were allowed to remote into their desktops. They were issued laptops that would be allowed to remote in to their desktop, and nothing else. You'd think that would be enough to send me running, but that was just the beginning. It's like they were collecting strange IT issues. They manually whitelisted all URLs on the firewall, so no sites worked since all trackers failed and all external media or anything embedded fails. They had a distro list for alarms instead of a mailbox, so I would get 1000+ alerts a day. My mailbox had 20K unread inbox emails on my first day. They "secured" the network by believing that having things on 10.x.x.x instead of 192.168.x.x somehow provided an additional level of security. The misconfigurations were causing millions of errors a day and broadcast storms. When I showed them the errors, they told me it was by design. Streaming anything over wifi was a term-worthy offense due to how often it crashed the network. I ended up leaving because the IT "manager" didn't believe I had been logging in enough remotely and decided I wasn't really working and just "looking for errors". They fired me, and they still think I was the problem.

Reply

[-]

simAlity@reddit (OP)

That sounds like Hell.

Reply

[-]

Landscape4737@reddit

Selling a server operating system that would cause it to fail and require a reboot every 49.7-days - windows nt 4 and some versions of windows 2000.

Reply

[-]

simAlity@reddit (OP)

Was this something that was really happening or just something that the office thought would happen?

Reply

[-]

jeffrey_f@reddit

Implementing the following at one company: Group policies on computers. Only AD computers on network Strong passwords and expiration dates (90 days). Some had the same password for years. Blocking spam. Killed almots 5K junk mails daily Moving data to protected (AD Groups) folder. Accounting, IT, Executives, etc No more users downloading and installing what they want. Admin only can install stuff No more local users with admin. AD users only That was the first few months.

Reply

[-]

tsuserwashere@reddit

Forcing frequent password changes tends to encourage poor password hygiene (lots of `SallyMaye!Q124`). As long as you have MFA and roll passwords if there is suspicion of a compromise, letting users keep passwords they’re comfortable with reduces security friction.

Reply

[-]

jeffrey_f@reddit

One company rewarded complex passwords with a much longer expiry. If you used a passphrase that was 20+ mixed alphanumeric + special characters with spaces, you had a year between password changes. If you were a minimalist, it was the standard 60-90 days

Reply

[-]

Vexser@reddit

I've read of old unix boxes hidden away in closets that nobody dares reboot because nobody knows anything about them and the software is unobtainable, the source code lost to the sands of time. And some of them might be doing critical functions which nobody understands. (I've seen this firsthand)

Reply

[-]

tekno45@reddit

we set some machines to static ips in aws because DNS went down once.

Reply

[-]

Landscape4737@reddit

Selling an operating system that needs rebooting every night - windows NT 4 and earlier when using terminal services.

Reply

[-]

Vistaer@reddit

Person going bananas on us for leaving a web browser open on their system to hp’s website after we downloaded their new printers drivers. “That’s how you let hackers in and viruses in!” “Yes it is, my nephew told me that he knows more than you!” My boss then had to apologize to her on my behalf after I started laughing so hard.

Reply

[-]

Doso777@reddit

Sysadmins asking users to give them their passwords for a cloud service.

Reply

[-]

Corgilicious@reddit

I’ve shared this before, but my organization actually recommends that people restart their machines each morning. Come to your desk, restart your machine, go get your coffee water or tea or whatever you need. Sometimes I would be speaking with a user and I would ask them if they had rebooted their machine recently. They would say oh of course you do it every morning. What they didn’t know is that we had a punk report that I could very easily run and pull all of the reboot events on the machine. Imagine they’re surprise when I would say well user, the system shows that you actually haven’t rebooted your machine in 15 days. They would always go very silent, and then say, “you can see?” Yes, yes I can. Now I will be more able to help you if you’re actually honest and accurate about the things that we share.

Reply

[-]

Sutar_Mekeg@reddit

If I send a ticket to the appropriate team to resolve, they just check the one alert that was used to generate the ticket, and if that one is clear, they close the ticket, dozens of related alerts be damned. As far as a cure, haven't found one yet.

Reply

[-]

onebit@reddit

Rebooting servers is bad luck.

Reply

[-]

No_Resolution_9252@reddit

Not patching because of "reasons" was the most prevalent. I couldn't even count the hours I spent troubleshooting something that had partially or entirely broken something because installing a patch, driver or firmware "might cause it to go down"

Reply

[-]

RedHal@reddit

"It's the network." No it isn't. It's the application that downloads three and a half thousand options for a dropdown box and the dev team refusing to implement something as basic as AJAX. (cue demo)

Reply

[-]

clubfungus@reddit

A belief that email addresses are case sensitive. In spite of repeated attempts to demonstrate, explain, and plead, I can't convince them otherwise.

Reply

[-]

Disastrous_Meal_4982@reddit

Wifi isn’t fast enough for thin clients. I just plugged one end of a Ethernet cable into the client and the other end was just hidden behind the desk and the “my system is slow” tickets stopped. Lasted for almost a year when one of the new techs noticed and removed all the cables. People got really upset that their machines were moved to WiFi without notice and I had to tell them that I had never plugged them in to the network to begin with. Saved the new guy, but they gave me crap until I quit a few years later.

Reply

[-]

elpollodiablox@reddit

I had a guy who was convinced the AP near him was giving him headaches, so he'd unplug it. Then he would unironically send in a ticket saying that wireless wasn't working properly. I finally took it off of the drop ceiling frame and mounted it on a girder above the ceiling panels (just directly above where it was previously mounted), but didn't tell him. When I saw him I'd ask him how his headaches were doing and he'd say he was just fine now that the AP wasn't where it was before, and wireless performance was much better.

Reply

[-]

Al0ysiusHWWW@reddit

Everything was port locked by MAC down to the port. Everything. Even our imaging station had to have MAC addresses sent and responded to before we could start meaningful work then again sent for their respective destinations including what machine they were replacing. Nightmare to do support also users move computers around all the time! They’d port lock on cable mix up and for whatever reason the only one who with permissions to address it was the network ARCHITECT who refused to directly respond to tickets.

Reply

[-]

XInsomniacX06@reddit

Had a client who would turn the power off to their shop with the breaker every night , took me a while to realize why they always had problems until I stayed late one night, and was just so surprised. 10 computers , networking equipment etc just cutoff every night

Reply

[-]

golfing_with_gandalf@reddit

Had one satellite office that started a rumor & ran with it that the wifi was causing migraines. Prior to 5G/COVID nonsense, even. We just ended up hiding the SSID names, told them we disconnected everything and asked them to report back and magically they felt better. Never heard about it after that. Another site-specific one was in construction, a certain project got into the habit of moving the wifi AP away from where IT set it up. They'd always move it and rearrange the office and not tell anyone, and what would happen is they'd drag it with the ethernet cable along the floor and just treat the cable like a power cord you could step on without problem. Explained it to them several times to stop and inevitably the cable would end up crunched into oblivion and their network dead.

Reply

[-]

Tatermen@reddit

For a long time - USB serial cables. Part of our business is installing and maintaining phone systems. Once upon a time, about 20 years ago, someone tried using the cheapest nastiest $5 USB serial cable to program a phone system. It didn't work because the cheap cable lacked the extra pins (CD, DCD, DTR etc), at which point USB serial cables were BANNED by the general manager. We spent probably a small fortune special-ordering laptops and computers with built-in serial ports for our telecoms engineers for about 15 years until having to maintain those old systems died out. And that's just one of many stupidly things that are banned because "we tried that once a decade ago and it didn't work right".

Reply

[-]

pdp10@reddit

I'd bet that more than half the market for laptops with non-USB serial ports, was from that type of occurrence.

Reply

[-]

thedanyes@reddit

Yes seems like throwing the baby out with the bathwater. Surely even 20 years ago there existed high quality USB-to-serial solutions fully-compliant with the RS232 spec.

Reply

[-]

Tatermen@reddit

There was. They just didn't want to spend the money on one. They'd rather spend thousands of dollars on something that definitely worked than risk 50 dollars on something that might.

Reply

[-]

Dangerous-Ad-170@reddit

Kinda reminds me of all the old timers I’ve seen who insist on using stock Cisco DB9 to RJ45 serial cables chained to a USB to DB9 adaptor. They just added an adaptor to the stock tool they already used, which makes sense in a way, but the USB to RJ45 console cables have been around for ages now and are just as cheap.

Reply

[-]

Bogus1989@reddit

once when about 30 people flew out to help with our emr go live,(we recently merged, and prior to that we never met anyone from outside our team and region in IT, and basically stopped all communication, since it was crickets) they didnt know our environment…well one hospital, one i originally was handed over and took from zero to hero and got down to so little tickets we didnt permanently keep someone there….after those guys left, we got a bunch of tickets? i kept coming upon devices id never seen before, stuff that was old as shit. stuff that was just ewaste. like nurse carts, that was originally a huge problem, i tossed all the faulty battery and fucked up ones years ago, eliminated all carts besides brand new ones (3 per department in case of a failure.)but no it was shit that didnt work…basically these end users pulled shit out of closets and made these unknowing IT guys fix it for them, and they got it up and running, but werent there for the failures….. soon as i recognized this, i stayed late one day and snuck every piece out to my truck, and hauled it all downtown to that dumpster, that no one knows about in loading dock. AINT havin NO CHANCE of those guys dumper diving having me do Frankenstein shit for a critical at 2am!

Reply

[-]

craniumcanyon@reddit

The finance secretary didn’t know how to send an attachment without using the “send to email” button on her all in one printer, so she would print a document, load it into her scanner, scan it on her all in one printer so she could then send it as an email attachment.

Reply

[-]

Phayzon@reddit

I had an old supervisor do the same thing. Showed him how to do it the right way and he said "I've been doing it this way for over a decade, and I retire next year. I'm not changing how I do things now." Fair enough!

Reply

[-]

stuckinPA@reddit

I had dental X-Ray computers and dental techs simply never thought of rebooting. I'd get a weird problem 2-3x a week and it always resolved with a reboot. so one day I created a scheduled task to reboot daily at 5AM. Now I get maybe two weird probs that required a reboot each year. No one cares that life is easier. No one correlates it to the systems are rebooted daily.

Reply

[-]

Commonpleas@reddit

Sadly, most wizardry remains invisible to the user.

Reply

[-]

RaNdomMSPPro@reddit

Patching is bad luck, or more accurately: if it’s not broke, don’t fix it.

Reply

[-]

billyyankNova@reddit

It wasn't completely office-wide, but one place I worked, quite a few execs used Outlook as a file server. Massive .pst's full of attached Excel and Word files. They had a personal share and department shares on the real file server, but somehow this made the most sense to them.

Reply

[-]

ender-_@reddit

I've got a client with a shared mail account being used as database. Only 334 GB, and they keep complaining that Outlook is freezing, but don't want to hear anything about using something more suitable.

Reply

[-]

QuantumWarrior@reddit

Last company I worked for had a similar behaviour, it kept them running an EOL Exchange server "just in case, as an archive" under direct order from the C-levels long after they'd been migrated to O365. Then an attacker got into the network and crypto'd most of three offices' data. Entry point was later found to be - surprise surprise - an old Exchange server vulnerability!

Reply

[-]

GildMyComments@reddit

What was your solution implementing the reboots? We have a bunch of azure-joined laptops that I need to make that happen on. I thought an azure script but worried it may hit and cause reboots on computers that just got turned on for the first time in a while.

Reply

[-]

simAlity@reddit (OP)

We were using KACE at the time. Microsoft pushed out updates twice a month and they always required restarts. I set KACE to force a restart within 48 hours of installing the updates.

Reply

[-]

MethanyJones@reddit

I may have perpetuated the belief that the token could fall out of the big IBM connector if you unplugged a workstation. It discouraged a few DIY moves.

Reply

[-]

a10-brrrt@reddit

That Dell computers would not work with HP servers.

Reply

[-]

Valheru78@reddit

I had it the other way around, since they were used to rebooting their Windows NT (yes long ago) servers every week they also had a policy to reboot the AS400 Unix machines every Monday morning. They couldn't fathom that this was not needed.

Reply

[-]

SibLiant@reddit

You'll find religious / magical thinking trying to scale the walls and invade the strongholds of science in every way possible.

Reply

[-]

PBRmy@reddit

Not necessarily "bad luck", but the computers were so slow and OS so bloated that users really had to consider if they had the time to reboot. Best case scenario you were looking at ~20 minutes to complete reboot and get back into the programs you need. You were also gambling on whether there might be some kind of update that would require to be applied which was completely unpredictable. It could be hours. If you had a meeting in an hour, you couldn't risk rebooting right now. I didnt cure the problem at all. Low level help desk at the time, no power to fix that problem. We'd just document that a user couldn't reboot now and would get back to us when they had time.

Reply

[-]

RCuber@reddit

Vista flashback, I had to switch to Linux.

Reply

[-]

pdp10@reddit

When reboots are legitimately painful, reboots happen less often. One of the things I appreciate most about desktop Linux are the low-impact updates. Update, whenever. Reboot, whenever. There's minimal causal effect between the two (but it's still easy to figure out whether a program or kernel has been updated on-disk, if you want to know).

Reply

[-]

Walbabyesser@reddit

Sounds like the steam powered ones of ye olde days ![gif](giphy|onnfUeQK0imVG|downsized)

Reply

[-]

hudsoncress@reddit

I don’t know what you’re talking about. Rebooting computers is definitely bad luck

Reply

[-]

Walbabyesser@reddit

https://preview.redd.it/3vam0mfugjfg1.jpeg?width=500&format=pjpg&auto=webp&s=33fcbb93ae295868716519121461e5b9446706ab

Reply

[-]

en-rob-deraj@reddit

I created a powershell script reminder that tells people the last time they rebooted if it exceeded 24 hours lol. Pops up daily until they reboot. Has worked great.

Reply

[-]

BloodFeastMan@reddit

One thing I'll throw in just for context, and this is actually kind of counter-intuitive, A "hard" reboot, i.e., you shut the computer down and re-start it later, saves the state of the machine to a degree, the assumption being, you're done working, and will turn it on again later when you want to resume. A "soft" boot, a "restart", does not do this, the assumption being that something either needs to be updated, or something's wrong, and so it boots to a clean slate. I know that doesn't really address OP's topic, but I thought I'd throw it out there.

Reply

[-]

electricheat@reddit

you can disable fast startup to remove this non-feature

Reply

[-]

Walbabyesser@reddit

Default setting for our domain

Reply

[-]

BloodFeastMan@reddit

Yeah, pushing stuff to users isn't really part of what I do, and it's not really on anyone else's to-do list. Personally, I restart my own machine each morning, makes me feel all warm and fuzzy. :)

Reply

[-]

sexbox360@reddit

Unrelated but I dislike users who have a reboot ritual. ie rebooting every morning. It sounds nice but they treat it like a magical spell that absolves them of all computer problems. They open tickets like "hey my outlook isn't working, I already rebooted" but when you press them on it, in reality they rebooted 8 hours earlier. You're supposed to reboot after problems occur. Not before.

Reply

[-]

Walbabyesser@reddit

Our software deployment runs at system start so rebooting from time to time isn‘t bad

Reply

[-]

binaryhextechdude@reddit

Well I never turn my office pc off. So I open everything Monday morning, run until Friday evening and then reboot unless I need to reboot for patching but either way I always reboot Friday night. Having it always on means I can remote in from home and access that totally not relevant to work file I was working on over the weekend lolz.

Reply

[-]

Calm-Bed4493@reddit

Yup, and it's always....4 days ago it rebooted

Reply

[-]

bbqwatermelon@reddit

My current org did not believe in updating server firmware for fear of introducing problems. I fixed that with a cadence using OME.

Reply

[-]

Hundike@reddit

I have seen a device with 700 days uptime. We were asked why it's slow? Then we've gone the other way, to turning everything off at night to save money and be more eco conscious. Like turning equipment off from the power button that should never be turned off and has stickers on it not to turn it off. Then it breaks and they call saying they don't know what they did wrong.. If only they could read?

Reply

[-]

glassmanjones@reddit

Puter comes with a certain number of boots. Every reboot uses one up. No different than replacing a crush washer some folk think.

Reply

[-]

Slottr@reddit

Twice monthly? Screw it I'd be forcing it weekly lol

Reply

[-]

simAlity@reddit (OP)

Nah. Twice monthly was enough to give windows what it needed without fully taking responsibility for restarts out of the user's hands. They have since acquired better habits and are now fairly easy to support.

Reply

[-]

CollegeFootballGood@reddit

Not too bad but we called everybody on the team by their initials. Kinda weird

Reply