Well at least we got a backup, right?
Right???
Post funny things about programming here! (Or just rant about your favourite programming language.)
Well at least we got a backup, right?
Right???
It last ran a week ago and we technically haven't tested it. Just our hot replicas which also just deleted all that data.
And of course by now every downstream system replicated AND CACHED that data.
This is what we in the industry refer to as a “big oof.”
I thing the technical term for this is an RGE.
(Resume Generating Event)
But it’s only, like, a handful of rows 🙃
It's a good way to wake yourself up in the morning
Doctors HATE this one simple trick! Lose up to 100% of MyChart data - and KEEP it off!
Can help reduce blood pressure, high cholesterol, weight, height, gender, name and more to NULL! Wake up feeling NULL and NULL!
For everyone's sanity, please restrict access to the prod DB to like two people. No company wants that to happen to them, and no developer wants to do that.
Me applying for any database access ever: “read only. I do not want write. READ ONLY.”
Datagrip has an option, and likely other database IDEs do as well - “Connect as READONLY”. Makes me feel a little safer
Just a funny story. All of our devs and even BAs used to have prod access. We all knew this was a bad idea and put in a process of hiring a DBA.
I think in the first two weeks the DBA screwed up prod twice. I can't remember the first mess up but the second he had a lock on the database and then went to lunch.
We eventually hired two awesome DBAs to replace that one but oh boy.
Imagine being hired to help prevent people from fucking something up, only to fuck that thing up in your first week—not once, but twice. You’d think after the first time it wouldn’t happen again…
8388409 = 2^23 - 199
I may have noticed this on a certain other aggregator site once upon a time, but I'm still none the wiser as to why.
199 rows kind of makes sense for whatever a legitimate query might have been, but if you're going to make up a number, why 2^23? Why subtract? Am I metaphorically barking up the wrong tree?
Is this merely a mistyping of 8388608 and it was supposed to be ±1 row? Still the wrong (B-)tree?
WHY DO I CARE
Are you Ramanujam reborn or a nerd who put every number they found on wolfram alpha?
In a place for programmer humour, you've got to expect there's at least one person who knows their powers of two. (Though I am missing a few these days).
As for considering me to be Ramanujan reborn, if there's any of Srinivasa in here, he's not been given a full deck to work with this time around and that's not very karmic of whichever deity or deities sent him back.
I know up to like 2^16 or maybe 2^17 while sufficiently caffeinated. Memorizing up to, or beyond, 2^23 is nerd award worthy.
Ramanujan reborn - the main protagonist from the Wheel of Maths books.
Ah reminds me of the time (back in the LAMP days) when I tried to apply this complicated equation that sales had come up with to our inventory database. This was one of those "just have the junior run it at midnight" type of shops. Anyway, I made a mistake and ended up exactly halving all inventory prices on production. See OP's picture for my face.
In retrospect, I'm thankful for that memory.
Ive had one of those moments. Where you fuck up so bad that your emotions wrap all the way around from panic, through fear, confusion, rage, dread and back to neutral, and you go 'Hmm..."
In T-SQL:
BEGIN TRANSACTION
{query to update/delete records}
(If the query returned the expected amount of affected rows)
COMMIT TRANSACTION
(If the query did not return the expected amount of affected rows)
ROLLBACK TRANSACTION
Note: I’ve been told before that this will lock the affected table(s) until the changes made are committed or rolled back, but after looking it up it looks like it depends on a lot of minor details. Just be careful if you use it in production.
This is now the correct database.
Rollback.
I don't understand environments that don't wrap things in transactions by default.
Especially since an update or delete without a where clause is considered valid.
I'm a data engineer that occasionally has to work in sql server, I use dbeaver and have our prod servers default to auto-wrap in transactions and I have to push a button and confirm I know it's prod before it commits changes there, it's great and has saved me when I accidentally had a script switch servers. For the sandbox server I don't have that on because the changes there don't matter except for testing, and we can always remake the thing from scratch in a few hours. I haven't had an oppsie yet and I hope to keep that streak
Legit have nightmares about this.
Just hit Ctrl+Z to Undo
You can also do this by forgetting a WHERE clause. I know this because I ruined a production database in my early years.
Always write your where before your insert, kids.
This is missing NSFW tag!
That's what backups are for.
Checking the backups... Ah yes, the backup done in August 2017.
Hello boss, I broke the company. I'll see myself out
If you don't have apt backups, that is a failure of the process, not yours.
Plot twist, they were also the one responsible for developing the backup process.
You should take it upon yourself to make regular backups in case you fuck up really bad. I had an intern that deleted everything on its fifth day, luckily l was automatically making backups two times a day, so it was fine.
Why would an intern be allowed anywhere near prod DB? Do you not have lower environments?
I actually screwed up twice on dev environment. Luckily the second case was salvageable without using data from an old backup(I wasn't given one that time) and I managed to sweep it up fast.
I started testing my queries super carefully after the first incident, but I was too tired once that I forgot to restrict the update scope for testing and screwed up again.
'content'
:)
Whenever SQL databases come up, I always think of this video from one of the greatest 'content' creators in history.
Transactions are your friend here
Begin transaction;
Then
Your sql here
Double/triple check the messages/console for results. Look good?
Commit;
Worried?
Rollback;
Just be sure to mind your transaction logs for long running queries and by all things holy be sure you’re not doing this to a live db with a ton of transactions since you’re basically pausing any updates until the commit or rollback on the affected tables
This makes it safer but like... don't run queries on production outside emergencies ever.
That transaction frame, depending on your specific DB, may cause severe performance side effects.
Look, the safe approach is to write it into something, PR it, get it reviewed, and then run it as part of a structured deployment process.
oopsie daisy moment
Looks like little bobby tables is at it again. (edit: for reference: https://xkcd.com/327/)
Edit #2: For lemmy app users: https://xkcd.com/327
And thanks to @Gestrid@lemmy.ca for the correction.
This is giving me PTSD
Had something like that happen to a local dev database (thankfully). A dev next me blurts out "how to I rollback an update in SQL server"? He was used to Oracle and how easy it is to rollback something. Had to explain that commit just happens in SQL server regardless of whether or not you put that commit line in.