As I Live and Learn

Wednesday, August 03, 2005

The World is Not Enough

I'm NOT a manager. It's times like tonight that are the exact reason I do not want to BE a manager. They can't pay me enough to put up with crap like this.



tintros said:
Rejected. Entirely too vague. Please resubmit when more detail is available, thank you for considering LiveJournal Press.

;) Details. C'mon. Give.

vosch_karanek said:
*hugs the ange petit* You will make it through... you always do. If you need to chat or vent, you have my number and I am at work all day and into the eve tonight until about noon tomorrow. *winks* I might be bored. *sticks tongue out*

trekker9er said:
So, we all knew it was going to be a late night on Tuesday. Work work work, yes, we've resigned ourselves to that until this Site Refresh release is done (this weekend... somehow, someway, sometime).

Tuesday was also the PPD final release and the Shadow site initial release. My co-worker, we'll call him R, was the release manager for that releases. We decided this on Monday because the person who was supposed to be release manager had never been one before. And while we wanted him to get the experience, we decided it would be more wasteful of people's time if he did with R backing him up instead of R just doing it. So R was assigned to do it so the other guy would have more time to work on other stuff.

Tuesday evening comes around. The PPD release is started. And Project Management informs us that the live database has to be imported to shadow site for the shadow site release. We, development, look at each other dumbfounded. PM explains the situation, we sigh and agree because the business logic makes sense. So begins problem one: importing a live database takes 3+ hours. Not knowing this was going to be necessary, the import was not started until 7pm. Which means, AT EARLIEST, testing the shadow site might start at 10, but mostly likely later.

Around 8pm, most people - not associated with either release - leave. I would have, but I decided since I was in the middle of something I wanted to finish, if I could, before I went home. And I figured it would be good if I hung around to back up R on the release, just in case. He had already asked me about something once. And, he had asked for information that was in the release script he was supposedly following. (That should have been my first clue, I wonder how I missed it. Or maybe I didn't, and that's what caused that little nagging inside of me that kept compelling me to stay when I didn't have to.)

The little annoyances began when R starting giving me a sort of play by play of what he was doing. I know how I release goes, I really didn't need to know. And if I had the desire to know how things were going, I'd ask. Though I tried to curb my curiosity (I failed once) because when I am doing a release I very much dislike to be bothered. But he was IMing me with "I just..." as if he was looking for my approval. Erg.

As the import of the database progressed, I was surprised to learn R had not done the PPD release in the correct order. Why? Apparently he just forgot about the first step. And of course that was one piece of updating he did not give me. Not that I should have gotten any, because I WAS NOT SUPPOSED TO BE PART OF EITHER RELEASE. *growling sigh* And so testing which was already in progress on things that were supposedly done, had to be started all over again. R cleaned up the problem, restarted all the affected sites (as more than just PPD was affected by the first step), and then testing started all over again.

Moving on. The project manager decides that we should just finish getting the sites up, and testing can all happen in the morning, so that people aren't up until ungodly hours because of this last minute database import. Sounds nice, doesn't it? Well, the problem is we committed to the client to have the site ready by the time they arrive at work in the morning. Which for our primary client is 6:30am. So crap, the testers have to stay up and test tonight, because we have to be sure we do not have to roll back the release.

Soon we find out R set the username and password wrong for the website, and was running script against the wrong database. The first scare was did he screw up the live site?! Whew, no, turned out he ran them against a non-existent database, so no harm done, except loss of time. Geez.

Another issue arises in the release, I help quash it quickly so R can continue on with the release, and begin finishing up my own work with the intention of going home. It was already some later than I had intended. And then I realize one of our testers had left, I do not know when. Apparently he had finished all the testing he had been assigned. *shrugs* Fine, if he did his work and there were no major issues found, I guess he could go home, eventhough testing is supposed to stay until the end of the release by taking other testing assignments away from other testers so everyone finishes faster. But the tester who left is our new hire, as in like a week and a half ago, and he just doesn't know how a lot of things work around here. (The project manager should have informed him, but oh well.)

And then the (more?) big stuff starts to hit: The project manager come over to R and I as asks if we plan to stay after the site is up. R immediately answers no. Which stuns me. At least one developer has to stay until the end of the release in case there is a problem that either needs to be fixed or calls for a rollback of the release. As the release manager, R is that person. Other developers can stay if they wish, or if it has been decided before the release they will be necessary. So this now means I am condemned to stay until the end of the release, simply because I had the stupid idea to stay late to finish some of my own work.

The hell.

I mention to the project manager development has to stay and why. I do not say who that developer has to be. As I'm trying to make sense of all that just occurred, I look around. The project manager is our newest, hired last month, who has no experience running a release on his own. Our other two managers, who do know their stuff, are gone. No testers are left in the office. Only one is still "working", but is doing it from home where we can't see how he's doing or help him. And to top it off, he is not actually a tester, but a volunteer from another "department" and so is not experienced testing. And then there is R and I. And all day I hadn't given a second thought to R doing the release because he's done releases before and these two were supposed to be generic releases. Yet here I am suddenly having to back him up on stuff he is supposed to know - check that, stuff he KNOWS - and worried about what he is doing because already he's messed up the order of things.

It was ridiculous! How did I end up the only experienced person on this turned hard release?!

Now I'm sitting here ready to say, "It's not my release, I'm going home". But if I do, I know (feel/believe/whatever) that R would take advantage of the project manager not knowing how things work and go home. And then the release would end up not getting completed because some stupid little thing will happen that the testers do not know about, and will have no one to ask about or have to fix a legitimate problem. And then the release would not be ready by the morning when the client goes in to use the site.

Around this point my Mom (who I had been IMing with and venting to) asks " So isn't this someone else's problem? Why are they leaving and you staying?"

My answer: *sighs* Because neither John nor Shawna are here.

John and Shawna are the other two managers.

Eventually the database import completes. I start packing up, but take my time just in case. Sure enough, R moves forward with the release and then is baffled about why things aren't working. Well duh, the answers were in the release script but he didn't even see them (or didn't understand them? Not sure which is worse). Fed up, I sit down and finish off the release myself. And, as predicted, just after I get the site up and go in to set up configuration stuff, I find a whole section of the site that will not work because of an error being thrown by the code. This is the type of thing a tester would come across and by stopped by (and in this particular case the kind of thing that would force a roll back of the release). I mention it to R, and move on with the configuration settings. A minute later I glance up and see he's not working, on anything. I ask him if he's looked at the error, and he says no, and basically that he didn't know he was supposed to be taking care of it.


I'd finally had it. The site was up, the configuration done. I went home.

11:45 pm. 14 hours after arriving at the office that morning.

I try to calm myself on the way, and am initially unsuccessful. Suddenly I'm aware I am grinding my back left teeth. I don't do that. I can't even think of one other instance in my life I've grinded my teeth. Immediately I try to think of something to take my mind off the awfulness, and once I find a completely non work related subject to focus on, am successful. But *$&##$*@!

Post a Comment

Spotlight Posts





This page is powered by Blogger. Isn't yours?