It's apparent SE doesn't care that non-JP servers are encountering 2002 issues

[Yahallo]

sry to burst your bubble, my current work is it related but i also worked as a coder for almost 8 years (changed careers after i felt that i wasted 3 years mastering ruby for it to fall flat) and there is no excuse on this word for when a dude with a open source packet sniffer is doing a better job than your entire networking team

It also sounds like this bug only occurs at rather extreme conditions since if it has been around since 1.0, nothing really triggered it until now. If a bug doesn't appear, it can be pretty difficult to find it.

[Krotoan]

sry to burst your bubble, my current work is it related but i also worked as a coder for almost 8 years (changed careers after i felt that i wasted 3 years mastering ruby for it to fall flat) and there is no excuse on this word for when a dude with a open source packet sniffer is doing a better job than your entire networking team

I love that the team is acknowledging and fixing the problem, yet the trolls continue to troll.

Do you want the "you were right" trophy in your profile? Or maybe everyone should get a "SE missed something" hat in game?

[Caurcas]

I love that the team is acknowledging and fixing the problem, yet the trolls continue to troll.

Do you want the "you were right" trophy in your profile? Or maybe everyone should get a "SE missed something" hat in game?

Yes we do.

[Tanis_Ebonhart]

This aged poorly after the latest news.

Delete this topic, TC, and apologize for even making it.

[VelKallor]

It also sounds like this bug only occurs at rather extreme conditions since if it has been around since 1.0, nothing really triggered it until now. If a bug doesn't appear, it can be pretty difficult to find it.

If we ever meet, remind me to tell you the story about a bug that brought down a multimillion dollar company network for three days.

What was it? A zero instead of a 1 in a line of code that hadnt been used in AGES. To quote him "The more complex a system is, the easier it is for one tiny error to wreak havoc"

[Yahallo]

If we ever meet, remind me to tell you the story abut a bug that brought down a multimillion dollar company network for three days.

What was it? A zero instead of a 1 in a line of code that hadnt been used in AGES. To quote him "The more complex a system is, the easier it is for one tiny error to wreak havoc"

Honestly, legacy code is the worst to deal with. Often times it is not even documented, and the people who wrote it are long gone.

[Hasrat]

I still go back to my leaky boat analogy. Boat sprung a leak, devs bailed like hell while working on a patch. Turns out, there was more than one leak. Turns out, there was even one nobody would've noticed if the boat weren't overloaded.

[Shibi]

I'm in NA, haven't changed anything about my network or PC. Since they took servers down for that 2002 patch, I haven't gotten ANY errors in queue. Multiple times going through 3-5k. If you're still having issues, I highly doubt it's the code,

Not specifically aimed at you, just that you are the first in this thread doing it... I am a bit surprised so many took Yoshi's word for his "iTs NOt us, iTs bEcAUsE yOu USe wiFI!" excuse, the packet sniffs were clear, and the error duplicated by many people. It was also a timable error, it happened on 15 minutes... and if it sadly happened when there was a customer induced DDOS on the lobby server, you got chucked to the back of the queue.

It happened to people in NA, it happened to people in all regions, and it happened on a daily basis.

Because one's own experience was 2002 free doesn't invalidate the facts everyone else is seeing it, and even people on ethernet cable. But now finally at least, SE admit, the fault was in their code.

It's been a huge frustration to log in at 6000+ and get down to 1000 over the course of a few hours, fight the random 15 minute 2002's all the way until around the 500 mark.. then not be able to connect and get booted to the end of the queue.

Let's hope it improves now.

[VelKallor]

Patience. PATIENCE.

Let the team get its job done, let them work and fix this. The communication from Yoshi and the team has been unequivocal and absolute in terms of telling us the who what where and most importantly WHY.

NO ONE knew that a line of code from 1.0 was waiting to pop up and say "Hey there remember me?". It could not have been prevented, it could not have been foreseen. Take a deep breath, count to ten and WAIT. There is nothing ANY of us can do right now, jumping up and down wont solve it. The steps they have taken are appropriate, and i will say this loudly.

THEY WANT THIS FIXED AS MUCH AS WE DO.

This was their penultimate expansion, the capper to a ten year story and God knows how much sweat, blood and ungodly hours they put in to make this the best they can. FF 14 is now in many ways their flagship, they want this to go smoothly, as would anyone.

This was NOT what they wanted, in any way shape or form. This is what they DIDNT want.

You know it.

I know it

Cancelling further sales? Cancelling ads? That wasnt at Yoshi's level, that was at CEO level, and you can bet your sweet bippy Yoshi would have had to front up to the CEO with all of this, and then sit down to hammer out a solution. This will cost them a LOT, not just in money terms either. I dont envy them one bit for the predicament they are in, and all the "I told you so's" in the world does NOTHING

Anyone can be a back seat driver..its when they say sure, heres the wheel, YOU drive...is when you realise you dont know as much as you think you did.

I counsel patience.

[Saefinn]

If we ever meet, remind me to tell you the story about a bug that brought down a multimillion dollar company network for three days.

What was it? A zero instead of a 1 in a line of code that hadnt been used in AGES. To quote him "The more complex a system is, the easier it is for one tiny error to wreak havoc"

Speaking as a developer...this feels so relatable. But this is the right of it.

But to add my own story to yours, which isn't as bad, but you reminded me of one of my own bugs.

One of our customers had an issue in their payment gateway. I found the fix required a lot of work and parts of it to be re-written. Bear in mind, in the development process there's a lot of different checks along the way to ensure the solution isn't flawed, the code isn't flawed and that it passes all testing, and testing can be thorough because people will try to break it.

Within a few days of going live it turned out that under very specific circumstances if somebody went to pay towards their product it would give them a full refund. It was in these circumstances that it was triggering a refund flag. It is just the nature of development.

And given the nature of the 2002 error, I can see why they'd suspect connection issues/packet loss and when the evidence showed contrary, they investigated it.

The level of detail and transparency is not something I am used to seeing companies give because so often they'll save face and provide less detail so it cannot be scrutinized or used against them. But it's an approach I appreciate, because it cuts the BS and means people can give more meaningful feedback.