Things you wish you could say (at work)


$BOSS and $Cow-orker:

Yes, there is a large difference between a source port and a destination port.

No, having a dodgy certificate on our web application server will not cause the app server to suddenly stop talking to a remote host.

No, that Windows command will not work on a Linux machine.

No, we didn’t change anything on our app server. The Nyetworks team did something to make it stop working.

Just because a Nyetworks idiot can access $remote_site from his laptop that is not proof our app server can also reach $remote_site.


Beautiful :slight_smile: Seems the nyetworks team is in dire need of a LART


Dear $cow-orker

What day is it?

You are asking for VPN access at the 11th hour.

If something goes wrong, YOU can sacrifice YOUR weekend and sit at the office doing software testing.

I’m NOT going to assist you beyond the creation of a vpn username and password.

Learn to manage yourself, you’re a grown up.


If they’re like the Nyetworks Team I’m part of, we’re really burned out on the stream of malformed/incomplete firewall requests we get.

  • Requesting FW rules between hosts on the same subnet. With no FW between them.
  • It’s the Firewall’s fault your service isn’t listening.
  • We added the server you want, but now you’re drip-feeding us a half-dozen more spread over a week.
  • Your firewall request is actually a Load Balancing request, which has a different process due to risks involved.
  • General missing data like IPs, Ports, etc.


If you continue to send email directly to me when it should be sent to our ticketing system, I’m going to start deleting your message unanswered.


We don’t need to do any more cabling or add to the chassis infrastructure. All we need is another 64 Tb of server RAM for the G9s and all of these little intermittent problems will sort themselves out. Then, we can actually decom a few blades or move them to another site where they can do some good.

I’m the new guy. The sensible answer doesn’t get a lot of airtime because it costs money. The problem is that these G9 blades were bought with too little RAM to begin with and are starved. It’s expensive, yes. Unfortunately, it’s money that should have been spent before. Open up your chequebook and the answers will just fall right out.


Do it. At the very least, reply saying that you will not take any action on their request unless it goes through the ticketing system.
Bypassing the ticketing system means there is no real audit trail of what has been requested, it bypasses any scheduling that is inherent in the ticketing system, and it relies on you remembering that stuff needs to be done. If you forget something then it will be you that gets into trouble, not the idiots bypassing the ticketing system.


I’ve been tasked to look at a publicly-accessible ticketing system, as we want our clients to use it, which means better efficiency on our part.

And I’m thinking of using the same for IT support requests. Currently we do have one, but it is in-house, meaning our clients cannot access it.

OSticket springs to mind. Anybody have some good suggestions? We don’t mind paying for it.


Have you taken a look at SpiceWorks?


We can’t even keep two doctors busy on any given day, let alone the occasional three doctor day. What in the name of all that is good and holy makes you think we can keep FOUR doctors busy? Yes I know one is training. But this is quite ridiculous…

Thank goodness I’m in surgery tomorrow. And I’ve loaded my schedule such that I should never have to leave that room except for lunch and to go to the bathroom.


“OK” is an abbreviation. “Okay” is a word. Learn the difference.


A hahaha hahaha. So you got caught out, and had to come in to the office then.

It never is a good idea to ask for VPN connectivity at the 11th hour as there’s other things that may go like the Swedish chef.

Hope you’ve learnt from this. Always ask in advance. Don’t ask at the eleventh hour.

Failure to plan ahead = planning to fail.


To whom it may concern,

If you’re going to plug a random network cable into a wall socket, please make sure that the other end is plugged into a PC, or not plugged in… not plugged into an IP phone that is already plugged into the adjacent socket.

FOADIAF. kthxbai

(That little network loop took three days to find, due to the switch disabling the affected port, then re-enabling it randomly.)


But but but I was jes’ tryin’ ta be helpful here, them data will just dribble out of ta netwohk cable


I used ServiceNow at my last job. It wasn’t bad. We’re using JIRA by Atlassian ( here, and I like it. Cloud-based, so I can access it from home without my work laptop, which has been useful on a few occasions, and there’s an app for SmartPhones. There’s a free trial if you want to check it out.


I’ve used ServiceNow as well. Remedy is also cloud based now. Both of those options are very expensive, though, which is why I brought up SpiceWorks.


You might actually be able to believe what your yearly survey says when you stop installing software that monitors every second of your team’s day, including their personal potty break time. Just how much trust do you think they’ll have for you to tell you what they really think when you do something like this?

And, as an aside, did you ever think to consider that this would have extremely negative effects on a team that has been doing this work for 15+ years?


It works because you said so isn’t really explaining the actual logic behind it. If you’re going to use a formula as an explanation piece to why the situation we’re experiencing is the way it is, you’re going to need more than that to gain support.


Inputs occur from 5 separate models and are processed using proprietary methods where they are entered into a central database. Then, a miracle happens. Processing continues after black box operations are completed which will take a variable and unknowable amount of time.


Clustering on vmware is a real pita, esp when it is your first time.