Hey, Root? How do you test your slow query ticket, again? I didn't bother reading the giant green "Testing notes:" box on the ticket. Yeah, could you explain it while I don't bother to listen and talk over you? Thanks.

And later:
Hey Root. I'm the DBA. Could you explain exactly what you're doing in this ticket, because i can't understand it. What are these new columns? Where is the new query? What are you doing? And why? Oh, the ticket? Yeah, I didn't bother to read it. There was too much text filled with things like implementation details, query optimization findings, overall benchmarking results, the purpose of the new columns, and i just couldn't care enough to read any of that. Yeah, I also don't know how to find the query it's running now. Yep, have complete access to the console and DB and query log. Still can't figure it out.

And later:

Hey Root. We pulled your urgent fix ticket from the release. You know, the one that SysOps and Data and even execs have been demanding? The one you finished three months ago? Yep, the problem is still taking down production every week or so, but we just can't verify that your fix is good enough. Even though the changes are pretty minimal, you've said it's 8x faster, and provided benchmark findings, we just ... don't know how to get the query it's running out of the code. or how check the query logs to find it. So. we just don't know if it's good enough.

Also, we goofed up when deploying and the testing database is gone, so now we can't test it since there are no records. Nevermind that you provided snippets to remedy exactly scenario in the ticket description you wrote three months ago.

And later:

Hey Root: Why did you take so long on this ticket? It has sat for so long now that someone else filed a ticket for it, with investigation findings. You know it's bringing down production, and it's kind of urgent. Maybe you should have prioritized it more, or written up better notes. You really need to communicate better. This is why we can't trust you to get things out.

*twitchy smile*

So my department is "integrating CI/CD"

Right now, there's a very anti-automation culture in the deployment process, and out of our many applications, almost none have automated testing. And my groups is the only one that uses feature branching - one of the few groups that uses branching at all beyond "master, dev"

So yeah... You could see how this is already ENTIRELY fucked from the very beginning.

First thing they want to do is add better support for a process... Which goes directly against CI/CD.

The process is that to deploy to production (even after it is manually approved by manager), someone in another department needs to press a button to manually deploy. This, as far as I can tell, is for business rule reasons rather than technical ones.

They want us to improve that (the system will stay exactly the same with some streamlined options for said button pressers)

I'm absolutely astounded at the way our management wants to do something but goes in exactly the opposite direction. It's like the found an article of what CI/CD was and then took notes on exactly what not to do.

Worst collaboration experience story?

I was not directly involved, it was a Delphi -> C# conversion of our customer returns application.

The dev manager was out to prove waterfall was the only development methodology that could make convert the monolith app to a lean, multi-tier, enterprise-worthy application.

Starting out with a team of 7 (3 devs, 2 dbas, team mgr, and the dev department mgr), they spent around 3 months designing, meetings, and more meetings. Armed with 50+ page specification Word document (not counting the countless Visio workflow diagrams and Microsoft Project timeline/ghantt charts), the team was ready to start coding.
The database design, workflow, and UI design (using Visio), was well done/thought out, but problems started on day one.
- Team mgr and Dev mgr split up the 3 devs, 1 dev wrote the database access library tier, 1 wrote the service tier, the other dev wrote the UI (I'll add this was the dev's first experience with WPF).
- Per the specification, all the layers wouldn't be integrated until all of them met the standards (unit tested, free from errors from VS's code analyzer, etc)
- By the time the devs where ready to code, the DBAs were already tasked with other projects, so the Returns app was prioritized to "when we get around to it"

Fast forward 6 months later, all the devs were 'done' coding, having very little/no communication with one another, then the integration. The service and database layers assumed different design patterns and different database relationships and the UI layer required functionality neither layers anticipated (ex. multi-users and the service maintaining some sort of state between them).

Those issues took about a month to work out, then the app began beta testing with real end users. App didn't make it 10 minutes before users gave up. Numerous UI logic errors, runtime errors, and overall app stability. Because the UI was so bad, the dev mgr brought in one of the web developers (she was pretty good at UI design). You might guess how useful someone is being dropped in on complex project , months after-the-fact and being told "Fix it!".

Couple of months of UI re-design and many other changes, the app was ready for beta testing.

In the mean time, the company hired a new customer service manager. When he saw the application, he rejected the app because he re-designed the entire returns process to be more efficient. The application UI was written to the exact step-by-step old returns process with little/no deviation.

With a tremendous amount of push-back (TL;DR), the dev mgr promised to change the app, but only after it was deployed into production (using "we can fix it later" excuse).

Still plagued with numerous bugs, the app was finally deployed. In attempts to save face, there was a company-wide party to celebrate the 'death' of the "old Delphi returns app" and the birth of the new. Cake, drinks, certificates of achievements for the devs, etc.

By the end of the project, the devs hated each other. Finger pointing, petty squabbles, out-right "FU!"s across the cube walls, etc. All the team members were re-assigned to other teams to separate them, leaving a single new hire to fix all the issues.

I fucking hate managers who go out spouting things have gone to shit and production is broken, when a single feature is missing from the application. It's your fault for not making sure the specifications were correct or not testing in QA before going to production! Just because one minor feature is missing does not mean production is broken you dimwit! It just means you suck at your job and we have to fix your fuck ups. It would also be really helpful if you morons accepted the fact that the teams want to do daily releases and continuous development, instead of having to deal with your insanity of pushing releases to a specific date so we have everything ready from marketing to Steve the janitor

Posted in DevOps discussion board (teams channel):

“Program x isn’t behaving the same way that it does on production. Can you please take a look?”

..a little background: we have a deployment scheduled for today and this issue was found during regression testing.

The issue found is that when a file is clicked on it disappears from the screen, and then isn’t opened…

The file is not on prem, and doesn’t get uploaded to a server that our DevOps team owns…

So why on earth would this development team be asking DevOps to look into a bug that is most likely a code related issue? 😆

Is this a common occurrence for anyone else?
A Bug is found, and the first thought is that the code isn’t the issue?

Best:
Seeing ALL the members of my team finally coming into their own. One person tackled our entire not-at-all-simple CI/CD setup from scratch knowing nothing about any of it and, while not without bumps in the road, did an excellent job overall (and then did the same for some other projects since he found himself being the SME). Two of my more junior people took on some difficult tasks that required them to design and build some tricky features from the ground-up, rather than me giving them a ton of guidance, design and even a start on the basic code early on (I just gave them some general descriptions of what I was looking for and then let them run with it). Again, not without some hiccups, but they ultimately delivered and learned a lot in the process and, I think, gained a new sense of self-confidence, which to me is the real win. And my other person handled some tricky high-level stuff that got him deep in the weeds of all the corporate procedures I'd normally shield them all from and did very well with it (and like the other person, wound up being an SME and doing it for some other projects after that). It took a while to get here, but I finally feel like I don't need to do all the really difficult stuff myself, I can count on them now, and they, I think, no longer feel like they're in over their heads if I throw something difficult at them.

Worst:
A few critical bugs slipped into production this year, with a few requiring some after-hours heroics to deal with (and, unfortunately, due to the timing, it all fell on me). Of course, that just tells us that next year we really need to focus on more robust automated testing (though, in reality, at least one of the issues almost certainly would not - COULD NOT - have been caught before-hand anyway, and that's probably true for more than just one of them). We had avoided major issues the previous three years we've been live, so this was unusual. Then again, it's in a way a symptom of success because with more users and more usage, both of which exploded this year, typically does come more issues discovered, so I guess it tempers the bad just a little bit.

Spent a couple of weeks on writing a cronjob which updates a certain value in the application config, and spend the last few months on testing it in different environments to make sure it does not fail in production. Ran the deployment script, and the damn cronjob fails because of ssl certificate on production. fuck me

Quick question for you all: How do you deal with a problem in production that you cannot fix, even over an extended period of time (say 2 months)?

For context I feel like I’m losing my sanity here, we’ve had this problem on our production API since the beginning of March this year. I’ve done so much testing, got in contact with various teams of my company to try to figure out any potential candidate that would explain the bug, but none worked out. No need to say I’ve spent a considerable amount of time searching on the internet for others with the same problem or similar… We’ve even opened a ticket with the cloud host to see if they would have more details about the problem without success. So how do you deal with that ?