What the most embarrassing technical mistake you ever made?

paul78

Some funny ones in here. thanks for sharing..

xxxooxxx

Not myself but candidates for jobs.

Back at my last job. We were recruiting a senior Microsoft Admin
Part of the interview was a quick 5 minute hand-on test.

We put a laptop on the desk and ask the person to join it to the domain (given domain, username, password). However, we only plug the network cable half way in.

About half of those MCSEs with 5+ yrs of experience failed the test.
"I think you have some problem with your DC"
"The hostname was already used"
"The laptop is already on the domain"
"The DHCP server was down, I could not get an IP"

It was priceless to watch their reaction, when they were shown the cable.

jamesp1983

Letter wrote: »

Had two putty sessions open at the same time, one to a core router and one to a test device... you can see where this is going.

ouch, that stinks.

N2IT

When I got back into IT in 2008 from a 3 year hiatus, I started reading a server 2000 book that was laying around. At this point I had 0 certifications and I didn't really didn't care for any to be honest. I always liked to read so I started reading up on DNS and DHCP etc. Well a DNS server was down and I was talking to a system admin guy and I kept saying DHCP server. I think he understood, but I felt like a dummy after the conversation was over and I thought about what I had said.

I looked like a guy who was trying to hard to be honest.


A coworker of mine was working a synch issue on a VP's my documents. Well the VP hadn't synched up for several months, there was a error pending and nothing beyond that error in the process would synch. It was all offline. So instead of having the customer copy the files to a folder on the desktop/C: OFFLINE, he just went ahead and blew the cache and resynched.

Talk about an uncomfortable sight. At one point we had a associate vp followed by a director of infrastructure and another IT director and level 3 manager his manager and his supervisor and my co worker at the end of the line. They were walking to SME to SME trying to find a way to recover the documents. It was ugly very ugly and I actually got up and went outside, this was a friend of mine. He felt terrible for weeks.

On top of that there was a contract in that file that was for a multimillion dollar deal.

DevilWAH

I know a guy who on a very very large customers network was working on mutiply Cisco devices.

he wiped the config on the 4 of them intending to copy a new one over, before rebooting them in stages.

Sadly he got confused and rebooted them before copying the new config, in one step he managed to take down both primary and secondary devices across two data centres. he managed in a second take out ever redundant system and take down a large chunk of the company network.

he no longer works for the company, he was dismissed when he could not give a good reason why he had removed the config from all devices at the same time, leaving the network in a such a vulnerably state. He would have been OK if he had just rebooted by accident, and the devices had come back. As it was the had to send out engineers to both location and had a down time of over an hour.

VAHokie56

REMOVED UNNECESSARY QUOTE OF PREVIOUS POST

whomp...whomp that sucks...

demonfurbie

when i was new to wiring i wired an entire location with crossover instead of patch

MAC_Addy

I once made an image to sit on our ghost server with the Registered To as "xx xx Pubic Schools" that's right, I put pubic, not Public. My coworker found this so amusing that he did a screen-shot and framed it for me. He's one of my best friends though, so it was all fun.

Mrock4

When I first was introduced to foundry (now brocade) network devices, I wasn't used to the CLI, and had all of 2 days to get used to it. I was SSH'd into a distribution switch at a nearby facility, when I learned that you can enable port-security on a foundry globally (which applies to trunks too). Needless to say, that day we had a glitch in the flux capacitor that caused a building to lose service until I could get onsite and fix it...

Also, when working on remote sites (during non-business hours), I make it a habit to use the "reload in XX" command, to ensure that if I get locked out due to a routing change or such, that the device will reboot and therefore void any changes I made and allow me back in. One time, I was on a voice gateway router, and did "reload in 20", then got pulled to go troubleshoot a fire for another customer. About 30 minutes later I realized I MAY (couldn't remember) have not cancelled the reload.

"Sh ver"
"uptime is 9 minutes"
"oops"

I don't like that command anymore.

DevilWAH

MAC_Addy wrote: »

I once made an image to sit on our ghost server with the Registered To as "xx xx Pubic Schools" that's right, I put pubic, not Public. My coworker found this so amusing that he did a screen-shot and framed it for me. He's one of my best friends though, so it was all fun.

Being dyslexic I have had a few of these moments, asking a very nice user who was very well spoken and in here 60 if she could "Cu&t and Paste" a screen shot for me in an email was one I remember. I don't know what it is but lots of simple words with minor slip ups seem to end up rather rude!

I have grown to be much more careful over the years.

Kelkin

I turned on debug ip packet on a cisco 7120 VPN router that had 100+ Ipsec/gre 3des vpn tunnels.. ever see a cpu on a router go from 20% to 100% within a second?

paulgswanson

Guys in my office couldnt ping a PC for whatever reason, They tried unplugging the cat5 and plugging it back in rebooting the machine reimaging it, new motherboard even... (I was ot involved in this, just to be clear) turns out...it was in the VoiP port that was clearly VERY CLEARLY labled...

paul78

paulgswanson wrote: »

turns out...it was in the VoiP port that was clearly...

It's funny how that old adage "did you plug it in" rings so true more often than we all care to admit. Good one.

psenior

The worst thing I did was shut down a server in the data center without following the change management process we had in place. The server admin gave me a verbal OK, so I went ahead and did it. However, this machine was very old and once I turned it off, it died and never came back on again. This created a big problem for the factory and the server admin and myself got deservedly chewed out by our manager.

jibbajabba

Shutdown the wrong server via command prompt (thinking its the local machine). Not a big deal usually as most server are connected via APC power bars with remote interface. So logged in, located the interface and powered on. Nothing happened, so reset the port and lost all connectivity to any remote card. As it turns out someone crossed the wires and I shutdown the switch handling all management traffic including the network of the admininterfaces of the APC bars (guess which topology got redesigned the very next day).

Problem was - the system in question was our VPN server I shut down so I could not work from home any more either. Needless to say - it was a 4hr round trip to the datacenter to push an on button.

Colleague hit a main breaker with his butt in a datacenter once. Powered off the whole floor - stupid design obviously but needless to say embarrassing.

jamesleecoleman

I was helping out a buddy with some 35XX switches. I was logged into the line console port and something happened. So I tried to log in again in another window but nothing would show up. So I was standing there thinking what was going on and I told my buddy. Well I had forgot that there was only one connection allowed using line console port.

Not really a good story.

drkat

Cisco ASA 5505 - remote end sending in ISAKMP requests.... not seeing anything in the debugs but seeing the packet captures but no drops..

well first I didnt have my debug level on lol..

Then ... well turns out we had a static nat on the asa for the outside interface (same interface trying to isakmp) and well... ... cleared it out and still didnt work so I figured that wasnt the issue as we had some other tunnels up...

escalated to my colleague who has more asa experience and boom.. fixed 5 minutes..

took the nat out and cleared connections/xlate <<--- forgot to clear the xlates. My piece of humble pie tasted sweet

vinbuck

Multiple STP botched config = switching loop....

Across three states

jmritenour

One time, I was upgrading a few ESXi 4.0 nodes to 4.1, and started VMotioning my VMs over. Customer was very specific about where it wanted VMs to go, so I was manually migrating them instead of just letting DRS evacuate them. Well, there was a DB server that for whatever reason took an hour (seriously) to vmotion, which ate up a good half of my 2 hour maintenance window. Just as it finished, I got ready to kick the host into maintenance mode, when I see it is starting to migrate back... because I forgot to disable DRS on the cluster.

breannaall

*sigh* wish I knew what half these things were

Novalith478

It's never too early to start learning.

Alif_Sadida_Ekin

At my previous job, I was SSH'd into a remote linux server located across the country and I had to create several mount points to our network shares. At the time, I thought the fstab file was interchangeable among machines, so I just copied it over from another server so I wouldn't have to rewrite everything. Apparently there are things in this file that are machine specific, and upon reboot I couldn't connect via terminal anymore.... When I told the senior administrator what I did, he said, "yeah you really really shouldn't have done that".

Luckily I had created a copy of the original fstab file and there was another admin onsite where that server was located so he was able to revert my mistake.

Hypntick

At least you thought to make a backup of the file. Could have been way worse.

RobertKaucher

Amen! When I'm training the noobs the first thing I teach them is make a copy of the file before you do ANYTHING. Web.config.05242012 is standard practice. Delete anything older than 3 months.

joshmadakor

I troubleshot failed PXE boot in my test lab for about an hour before I realized all the lab's network infrastructure was powered off

Darthn3ss

rm -rf / tmp/asdfsa

Wasn't me, but someone posted it in a similar thread on reddit.

RobertKaucher

joshmadakor wrote: »

I troubleshot failed PXE boot in my test lab for about an hour before I realized all the lab's network infrastructure was powered off

That's just a silly daily occurrence! You need to either share something serious or go mess something up!

DevilWAH

When I was much yonger our family got there first PC, one of the origianl IBM's with two 5inch floppies and a monocrome screen. IT al so had 2 X 10mb hard drives with DOS installed on drive c:

Quite soon after we got it I did the infomous del c:/*.*

If I hadn't known before I soon lernt that that stops a PC booting, and with out the internet to help or any books I had to get the thing backup and working before the rest of the famiy found out.

I think that one mestake taught me a lot about "knowing waht you are doing before you do it". It's deffently stops me making the same mestake on a work's network.

paul78

I just spend the last 6 hours trying to configure the sound drivers on my Linux host so that I could get sound to work on a VM of Linux Mint that I was trying out.

I must have recompiled the host kernel at least 4 times. And I even started to download the ALSA utils so I could compile it and try to step through the code to see why the drivers wasnt working.

As I was calling it quits for the night, I noticed that I had pluged in the speakers into the microphone plug - sigh -

Zorodzai

Way back in my days of tech support. Was doing routine maintenance at a client site (defrag drives, free up space, check av updates etc) and, for some stupid reason, ran a disk cleaner utility (CCleaner) without first asking the user if she saved any files to her temporary folders.................managed to clean up about 2 GB of her music files