Error 522 on low and talk

bikegremlin

I clicked on a "thanks" button and got blocked, immediately.

Is DDOS attack more dangerous than DWindows attack?

willK

Haven't been able to access LET for days. Hope it has nothing to do with Dorian.

angstrom

@bikegremlin said:
Is DDOS attack more dangerous than DWindows attack?

It would be nice if CC explained these things to us.

(For what it's worth, I've heard that DLinux attacks are the worst but they're very rare.)

Radi

LET's go on Reddit until LET is stable.

bikegremlin

@Radi said:
LET's go on Reddit until LET is stable.

... ten years later...

sidewinder

I can't connect from my home I.p.

Have to use mobile connection

Cloudflare is useless. Long live slowendtalk.com

fLoo

@Radi said:
LET's go on Reddit until LET is stable.

Please dont use Reddit. Those masochistic chinese payed fucking ops who delete and ban every board which comes along. I prefer this place as it is. Long live the oldschool message boards

sin

I've been getting a "Banned/Rate limited" error by Cloudflare when trying to visit LET the past day or so.

raindog308

It was like the LowEndApocalypse.

sidewinder

Icons don't work for me on the home ip now, but the site is reachable.

Cloudfare is good at making things worse for sure.

jsg

Wow, I just had error 1015 again.

LET said:
Error 1015
You are being rate limited
The owner of the website has temporarily banned you

(served by CloudF#%&!)

What evil things did I do?

• I logged in
• I scanned through the thread list and clicked (new tab) some 5 or so threads I was interested in (just as I usually do since I'm here)
• I looked at the first tab/thread and wanted to answer to someone - but I couldn't.
• Smelling a rat I clicked through the other tabs. Only one other tab/thread had opened. The others showed the 1015

Is this a joke? I mean, opening a couple of threads is in no way evil or uncommon. It seems that some CC admin has gone way too far with rate limiting. Well noted, I can and do understand that an unnerved admin uses rate limiting after a DDOS attack, but those admins should also understand that LET doesn't lend itself well to rate limiting due to pretty much every page being (over?)loaded with images which means that with http 1.1 a perfectly normal page load will make dozens of requests for all the gadgets, scripts, avatar images, etc.

alvin

LET becoming 522 daily fun..

Janevski

@TimboJones said:
This is like finding out your nutritionist eats worse than you or your house cleaners house is way dirtier than your own place. Or the guys who make your backup software don't take backups themselves.

Or the dentist, can't fix his own teeth.

hanoi

if LET need my $5 vps, i'm willing to donate. Otherwise I will keep holding F5 to DDOS.
Come on guys, we are all sysassmin here, please post your postmortem. We have right to know.

bikegremlin

@hanoi said:
Come on guys, we are all sysassmin here, please post your postmortem. We have right to know.

Isn't it obvious, with all the suspense, couldn't you figure out already that the culprit is, as usually, the butler!?

sin

jsg said: Wow, I just had error 1015 again.

Yup, shit is annoying as hell. I simply refreshed the page and got hit with that and then couldn't access LET for like 2 days (not to mention LET was slow as fuck for a week before that).

ITLabs

@hanoi said:
Come on guys, we are all sysassmin here, please post your postmortem. We have right to know.

Come on ... they didn't even realize that Low and Talk had been ddosed.

jbiloh

Hi Guys,

Uptime is very important to us and I regret that the site experienced a prolonged period of reduced availability. We've gotten quite good over the years, and especially recently, at minimizing the downtime on the sites despite near constant attacks on its back end.

Here is a snip from our engineers about the specifics:

Beginning Thursday an attack of lowendtalk.com began. Beginning Friday morning support were working with Cloudflare to find the source of the attack and neutralize it. Over Saturday and Sunday we went back and forth with Cloudflare before ultimately deciding to use rate limiting to thwart the distributed attack. By Sunday night the fix was in place and the attack had been blocked thoroughly enough that the site was usable. The following Monday we began seeing reports that legitimate users were being blocked. Yesterday, we upped the threshold for a block and were still seeing legitimate users being blocked. Today the threshold was raised again.

Let me know if you have any questions.

As always, thanks guys

drserver

It seems more like DB server had issues than DDoS attack. Rate limit will reduce number of calls what single IP is doing. In case of large scale attack this is something that cannot help you. On minor localised attack DNS filtering can help. Looks like you are referring to L7 attack which you can kill with simple captcha challenge and cloudflare. As i saw browser check enabled which sometime triggers captcha this leads to conclusion that DB server had issues.

jsg

@jbiloh said:
... I regret that the site experienced a prolonged period of reduced availability ...

Thanks for your post, sincerly. I value it because it nourishes hopes that LET actually means something, anything to you/the company.

But frankly, calling a massive multi-day virtually complete outage "reduced availability" seems to be a new peak in euphemisms. It's as if your mother were in a coma and you described that as her showing "reduced agility".

But again, after all the things I've read about you and the company I appreciate the fact that you wrote that post. Thanks.

JohnMiller92

@jbiloh said:
Hi Guys,

Uptime is very important to us and I regret that the site experienced a prolonged period of reduced availability. We've gotten quite good over the years, and especially recently, at minimizing the downtime on the sites despite near constant attacks on its back end.

Here is a snip from our engineers about the specifics:

Beginning Thursday an attack of lowendtalk.com began. Beginning Friday morning support were working with Cloudflare to find the source of the attack and neutralize it. Over Saturday and Sunday we went back and forth with Cloudflare before ultimately deciding to use rate limiting to thwart the distributed attack. By Sunday night the fix was in place and the attack had been blocked thoroughly enough that the site was usable. The following Monday we began seeing reports that legitimate users were being blocked. Yesterday, we upped the threshold for a block and were still seeing legitimate users being blocked. Today the threshold was raised again.

Let me know if you have any questions.

As always, thanks guys

Nice, thanks for the update.

I always thought CF was pretty decent for DDoS protection, however, I guess it was a rather large attack?

mountainpup

@jbiloh said:
Hi Guys,

Uptime is very important to us and I regret that the site experienced a prolonged period of reduced availability. We've gotten quite good over the years, and especially recently, at minimizing the downtime on the sites despite near constant attacks on its back end.

Here is a snip from our engineers about the specifics:

Beginning Thursday an attack of lowendtalk.com began. Beginning Friday morning support were working with Cloudflare to find the source of the attack and neutralize it. Over Saturday and Sunday we went back and forth with Cloudflare before ultimately deciding to use rate limiting to thwart the distributed attack. By Sunday night the fix was in place and the attack had been blocked thoroughly enough that the site was usable. The following Monday we began seeing reports that legitimate users were being blocked. Yesterday, we upped the threshold for a block and were still seeing legitimate users being blocked. Today the threshold was raised again.

Let me know if you have any questions.

As always, thanks guys

LMAO I always thought Cloudflare is very good at defining attacks mitigating company guess I was wrong.

They could barely mitigate the attacks on a niche website god help u if its state sponsored attacks, cf is all talk. I'm a customer shame.

I guess my imperva and legacy pzeroshield is better less bullcrap :P

stefeman

@mountainpup said:

@jbiloh said:
Hi Guys,

Uptime is very important to us and I regret that the site experienced a prolonged period of reduced availability. We've gotten quite good over the years, and especially recently, at minimizing the downtime on the sites despite near constant attacks on its back end.

Here is a snip from our engineers about the specifics:

Beginning Thursday an attack of lowendtalk.com began. Beginning Friday morning support were working with Cloudflare to find the source of the attack and neutralize it. Over Saturday and Sunday we went back and forth with Cloudflare before ultimately deciding to use rate limiting to thwart the distributed attack. By Sunday night the fix was in place and the attack had been blocked thoroughly enough that the site was usable. The following Monday we began seeing reports that legitimate users were being blocked. Yesterday, we upped the threshold for a block and were still seeing legitimate users being blocked. Today the threshold was raised again.

Let me know if you have any questions.

As always, thanks guys

LMAO I always thought Cloudflare is very good at defining attacks mitigating company guess I was wrong.

They could barely mitigate the attacks on a niche website god help u if its state sponsored attacks, cf is all talk. I'm a customer shame.

I guess my imperva and legacy pzeroshield is better less bullcrap :P

The amount of dedication you get depends on ur cloudflare plan. If you pay 5k USD for enterprise, you can be sure its their first priority to tune the service for you.

Theyre actually really solid, when you have access to a rep and technical team instantly.

ITLabs

It seems that "reduced availability" is (was) beginning to begin (began) again.

vyas11

@ITLabs said:
It seems that "reduced availability" is (was) beginning to begin (began) again.

Yes, on my mobile network and home internet. Need a detox. Or a strong cup of coffee. Or a self ban from visiting LET till BF.

mrclown

welcome back 522.

rcxb

@ITLabs said:
It seems that "reduced availability" is (was) beginning to begin (began) again.

You must be a technical writer...

"Cable Management: Can be mounted and dismounted easily. So many holes can satisfy with your different position and type request."

AnthonySmith

jsg said: It's as if your mother were in a coma and you described that as her showing "reduced agility".

fucking brilliant!

ITLabs

And here we go again... 522Talks.

a3Bren

today still 522.