IT/Software career thread: Invert binary trees for dollars.

Khane

Got something right about marriage
19,836
13,353
We use Wagile. Which is waterfall + agile and basically not a real methodology at all that takes all the worst parts of both and stuffs them into quarterly, company wide releases.

Every thing any team works on in any given period of time is going to be part of the "March release" or "July Release" etc. Needless to say release weekends are always really fun...
 

Noodleface

A Mod Real Quick
37,961
14,508
We have 3 week sprints and do big releases every 6 weeks (with intermittent releases on week 3).

I've never heard of a ten day sprint before. That almost feels like overhead would be at a point where it's taking almost as much time.
 

ShakyJake

<Donor>
7,629
19,259
It's 9 work days. Begins on Wednesday, ends on the Tuesday after next. On that particular Wednesday we have the Sprint Review / Presentation to stakeholders, then sprint planning afterwards.
 

Mist

Eeyore Enthusiast
<Gold Donor>
30,413
22,203
We use Wagile. Which is waterfall + agile and basically not a real methodology at all that takes all the worst parts of both and stuffs them into quarterly, company wide releases.

Every thing any team works on in any given period of time is going to be part of the "March release" or "July Release" etc. Needless to say release weekends are always really fun...
lol what
 

Mist

Eeyore Enthusiast
<Gold Donor>
30,413
22,203
Here's a good one.

We monitor/manage over 100,000 devices. Switches, routers, firewalls, servers, voice gateways, DC blades/chassis, etc for about 2000 customers.

All of these these need passwords, obviously.

We had a pretty unsecure system for storing these, and we were generally missing about 10-15% of the passwords for the devices, which was a real pain, but for the rest, everything was fairly well organized, just insecure.

So we purchased a product to store passwords securely.

About 10% of the way into migrating the passwords over to this product, we realized the database search performance was awful and that searches chronically timed out and returned 0 results. We also had no plan for migrating the data in an organized manner, so much of it is just indexed really badly, like "Router 1, Router 2, Router 3, etc instead of "Dallas Core Router, Dallas Backup Router, Chicago Core Router, Chicago Backup Router." We threw a bunch of hardware at the product to try to improve performance but it did nothing.

Despite all these problems, the migrations continue. Passwords go in, but no one can retrieve them. So we've basically ransomwared our own passwords, which is then causing everyone to just open tickets with the OEMs and get the passwords reset, which is then invalidating all the passwords that were migrated into the system, and they still haven't come up with any kind of documentation guide for how to properly put passwords in the database... so the reset passwords aren't going into the database.

This has been going on for like 18 months now and no one has a plan to fix it.

And the best part is that we've realized that the third party company in India that we hired to do our tier 0 event management triage a) keeps locking out all the accounts because they log in with the wrong passwords and b) when they do successfully log in, they put the passwords into their own account notes for speedier access, which totally negates any security we "gained" in this whole process.
 

Noodleface

A Mod Real Quick
37,961
14,508
A lot of companies are like this

Have you ever had someone ask for a recommendation, so you talk for 3 seconds and they go 'wow thanks!! Your the biggest help' and cut you off with this weird half ass advice

That's how most companies do dev methodologies
 
  • 1Like
Reactions: 1 user

alavaz

Trakanon Raider
2,001
713
It's just like anything else in the tech field. Our new dev shop manager decided that we needed to be "DevOps" which to him means "put shit in containers." We ask, what are we gaining by putting shit in containers? His answer "PUT SHIT IN CONTAINERS!"
 
  • 2Worf
  • 1Like
Reactions: 2 users

TJT

Mr. Poopybutthole
<Gold Donor>
40,949
102,793
It's just like anything else in the tech field. Our new dev shop manager decided that we needed to be "DevOps" which to him means "put shit in containers." We ask, what are we gaining by putting shit in containers? His answer "PUT SHIT IN CONTAINERS!"

I get that everyone thinks it's cool and all so they think they'll be a top tier shop if they use it.

It's so bizarre to be to see career devs with like a decade or more experience using tech and applying it not stepping back and saying, "ok but what does this do exactly?"

So you get all these projects where one dev came in and said such and such framework is badass let's make it the foundation of our project. Five years later that guy is gone and his team barely understood said framework now you're dependent on it without the ability to maintain it short of starting at ground zero again.

Like over half of all my projects at GM were like that.
 

alavaz

Trakanon Raider
2,001
713
I'm a container hipster in that I was using them way before they were popular (Solaris 10). I know many situations for which they are perfect for and a back office .Net web app doesn't really fit that bill. We do have things for which containerization would be awesome ironically, but none of which are under the dev managers purview.
 

Deathwing

<Bronze Donator>
16,395
7,395
Does that web app require a lot of configuration? Containers, and specifically yml files, are great for enforcing reliable configuration regardless of where it runs. Hardware dies? IT just needs to hand you a new machine that runs docker, that's it.
 

Noodleface

A Mod Real Quick
37,961
14,508
I rejected a code review today and heard the guy talking shit about me to someone else. Not the first time he's done it either
 

alavaz

Trakanon Raider
2,001
713
Does that web app require a lot of configuration? Containers, and specifically yml files, are great for enforcing reliable configuration regardless of where it runs. Hardware dies? IT just needs to hand you a new machine that runs docker, that's it.

No that's the thing, they use TFS and it automates the deployment to app servers. If the hardware dies, I just hand them a new app server that runs IIS and the .net framework and that's it. Containerizing these apps won't make this process faster, less complicated or more efficient, it's the same shit either way.
 

Lendarios

Trump's Staff
<Gold Donor>
19,360
-17,424
I rejected a code review today and heard the guy talking shit about me to someone else. Not the first time he's done it either
1) Punch him on the phase.
2) Talk to your manager about this.
3) Talk to him about it.
4) Do nothing.

pick one or 2 or 3 =) Talk to him, then punch him, then talk to your manager.
 

Noodleface

A Mod Real Quick
37,961
14,508
1) Punch him on the phase.
2) Talk to your manager about this.
3) Talk to him about it.
4) Do nothing.

pick one or 2 or 3 =) Talk to him, then punch him, then talk to your manager.
Boss is kind of tired of him. I think he has autism so he kind of grates on everyone
 

Deathwing

<Bronze Donator>
16,395
7,395
1) Punch him on the phase.
2) Talk to your manager about this.
3) Talk to him about it.
4) Do nothing.

pick one or 2 or 3 =) Talk to him, then punch him, then talk to your manager.
Amod Amod can we move every thread to the grown-up forum for a month so we can infract len for this?
 

Ao-

¯\_(ツ)_/¯
<WoW Guild Officer>
7,879
507
Here's a good one.

We monitor/manage over 100,000 devices. Switches, routers, firewalls, servers, voice gateways, DC blades/chassis, etc for about 2000 customers.

All of these these need passwords, obviously.

We had a pretty unsecure system for storing these, and we were generally missing about 10-15% of the passwords for the devices, which was a real pain, but for the rest, everything was fairly well organized, just insecure.

So we purchased a product to store passwords securely.

About 10% of the way into migrating the passwords over to this product, we realized the database search performance was awful and that searches chronically timed out and returned 0 results. We also had no plan for migrating the data in an organized manner, so much of it is just indexed really badly, like "Router 1, Router 2, Router 3, etc instead of "Dallas Core Router, Dallas Backup Router, Chicago Core Router, Chicago Backup Router." We threw a bunch of hardware at the product to try to improve performance but it did nothing.

Despite all these problems, the migrations continue. Passwords go in, but no one can retrieve them. So we've basically ransomwared our own passwords, which is then causing everyone to just open tickets with the OEMs and get the passwords reset, which is then invalidating all the passwords that were migrated into the system, and they still haven't come up with any kind of documentation guide for how to properly put passwords in the database... so the reset passwords aren't going into the database.

This has been going on for like 18 months now and no one has a plan to fix it.

And the best part is that we've realized that the third party company in India that we hired to do our tier 0 event management triage a) keeps locking out all the accounts because they log in with the wrong passwords and b) when they do successfully log in, they put the passwords into their own account notes for speedier access, which totally negates any security we "gained" in this whole process.
What fucking product is this? CyberArk works for x120k assets for us...