Linode Manager and API are under attack

https://status.linode.com/incidents/6mpxv406bhq9

by mherrmann • 10 years ago

68 39 10 years ago

status.linode.com

hnarayanan• 10 years ago

This sort of headline involving Linode is (unfortunately) beginning to feel extremely frequent. I am happy that I've since moved away to GCP, but this is still sad to hear.

charlieegan3• 10 years ago

Certainly often seems to come up on HN. https://news.ycombinator.com/from?site=linode.com

cstrat• 10 years ago

I feel bad for them and their customers.

Wonder what the root cause of all this is, they seem to be a specific target for some reason...

boulos2• 10 years ago

High value (a decent number of high profile customers) and yet easier than say us (Google), AWS or Azure. As the last discussion brought up, I'm curious whether Linode is particularly singled out compared to say Digital Ocean, OVH, Hetzner, et al. but maybe someone like jgrahamc knows?

asciihacker• 10 years ago

I am happy that I've since moved away to GCP

What is "GCP"?

jeffcore• 10 years ago

I believe they're referring to Google Cloud Platform.

jupake• 10 years ago

I'd be interested to know which data centres are being attacked. And why just the api and manager?

boulos2• 10 years ago

I don't know where, but as to the why: control planes usually have fairly low traffic (create an instance isn't that frequent per customer) so they'd be easier to swamp if you can get it to consider your traffic. Proper load shedding and rate limiting is kind of hard (you have to decide what you want to happen to well intentioned folks), but anyway that's probably why.

andruby• 10 years ago

Maybe this attack is part of a plan to get access to the control panel of customers. That would be a way to gain access to some relatively high profile sites.

user5994461• 10 years ago

And why just the api and manager?

Likely because it's more effective, if you really want to kill the site.

Let's say that you find a "/manage/viewall" page which takes some processing to load (more so than the other pages), is never cached, and can't be protected by cloudflare/capcha/other because it must stay open for CLI tools.

First, it's easy to overload the site by throwing a couple of expensive requests to a few vulnerable services, rather than to have a botnet flooding 100 Gb/s of traffic to random customer instances. (note: the two strategies are not mutually exclusive).

Second, by attacking these services, they impact Linode itself and all customers using it. Someone is really trying to hurt Linode by doing that.

The lesson here: There seem to be some naughty attackers who are putting a lot of effort to put Linode down recently... and they seem to succeed to some extent. That really is a bad position for Linode :(

JimDabell• 10 years ago

And why just the api and manager?

They are ancient Coldfusion implementations that have a dire security track record, so I would assume that they are the weakest link in the chain. Linode said they were rewriting them a while back, but I don't know what came of that because nothing seems to have changed in years.

deftnerd2• 10 years ago

My theory is either a ruthless competitor has been trying to destroy linode for years, this is a distraction to cover a different attack on them, or at some point in the past someone attacked and blackmailed them and Linode paid to make it stop. Once you set that president, it's hard to make it stop.

jacquesm2• 10 years ago

It's 'precedent' and without any evidence these theories are mostly worthless.

deftnerd2• 10 years ago

You're correct in both regards. As for the first point, that was just an auto correct mistake. My apologies.

On the second point, you're correct that my theories are entirely unsubstantiated. I'm just trying to guess why Linode seems to be the victim of these kinds of attacks a lot more than other providers.

sintaxi• 10 years ago

It's more likely a customer of Linode who is the target.

wepple• 10 years ago

You don't think a direct attack against the customer would be better spent energy than hitting Linodes APIs?

alex_hitchins2• 10 years ago

Out of interest, is there anything to suggest this is an outside attack or an inside attack?

viraptor• 10 years ago

outside.

1. "We are currently experiencing a DDoS attack targeting our infrastructure." - you don't call inside attack a DDoS.

2. For internal attacks they can just firewall off / kill offending servers. It would be resolved in minutes rather than "mitigated".

alex_hitchins2• 10 years ago

Thanks, I suppose I should have figure this out. I wondered if there was some attack that could be kicked of within the network that could be more difficult to trace. I guess you could just looks at port traffic at switch level.

xarope• 10 years ago

I feel bad for Linode. They've always been a good host (I use Linode, as well as Digital Ocean and others), so I would hate to have to switch away.

prawn• 10 years ago

Same. I've found their service to be great and even today they went out of their way to be especially helpful to me with a support query.

Can only imagine how much money and time is spent covering for malicious acts like this, malware, viruses, etc.

JshWright• 10 years ago

They have been a good host as long as you're not overly concerned with security (and more importantly, transparency in responses to security incidents).

endijs2• 10 years ago

I'm not really sure why "Linode" and "DDOS" makes the news. It happens regularly. And is nothing to be surprised about. What is new is how they managed to get their systems upgraded to the point where DDOS does not matter anymore. After big blackout (when they were under attack for weeks and even one of their DC's was completely offline for more than a day) their defenses have become way better. Of course - maybe attacks are not so big. I do not know that. It's my 8th year with them. During those years only once (I think) I had downtime because of those DDOS attacks. Not bad at all.

matt_wulfeck• 10 years ago

It's a huge deal for a linode customer. It's not like your site is DDoS'd and you can't access. You can't access anything on linode when their entire service is DDoS'd.

Part of the reason I think it's news is to shame linode. The likelihood of a DDoS attack bringing down the AWS management console is extremely low (has it ever happened?) Yet attackers have managed to do it multiple times on Linode.

endijs2• 10 years ago

I'm one of those customers (for 8th year already) and for me it's not a big deal. Unless you are one of those high profile customers who needs constant API access, those few minutes of downtime is not a big deal. Not everyone is spinning up and down new instances all the time. Plus - not all attacks result in downtime for them. Sometimes you do not notice at all, sometimes responsiveness will suffer a bit. And as I said - their defenses have become way better than they were a year ago. And it's not really fair to compare Linode to AWS. It would be more interesting to see how would DigitalOcean handle the same amount of attacks. And I doubt DO would handle them any better than Linode.

oliwarner• 10 years ago

Speaking as a customer, I've never actually seen any direct service downtime as a result of a DDOS. Specific services (usually their web manager) have gone down but that hasn't affected my clients' sites running on my servers.

Obvious scope for improvement —and I've no doubt this is a constant engineering sink— but not being able to commission new nodes through the web interface for a few hours hasn't yet ruined my day.

throw2016• 10 years ago

It feels off that Linode is being targetted so intensely and one feels bad for them but this can destroy their business.

It will be a pity if the barriers to entry are being constantly raised by the need for more and more mitigation and this will reflect in the level of expertise needed and prices for end consumers, and even then with the constant threat of blackouts.

It also becomes easy to silence inconvenient voices.

It is difficult to run a business or website if you need to constantly fight extortionists or make payoffs to address threats of downtime. Today Linode is in the news, tomorrow it could be anyone else targetted.

There is no easy solution here, without putting constraints somewhere but there must be a way to make the web more resilient and robust and not subject to the whims of extortionists and other malicious agents.

alvil• 10 years ago

Qui bono? :)

rebootthesystem2• 10 years ago

Linode customer here. Going to take a tangent.

Way back when, Toyota instituted a policy that required every line worker to effectively stop the factory if a defect was found. Engineers would come to the floor. They were tasked with creating a permanent solution for the problem, one that would have that never happen again.

For the first few months Toyota had a very hard time getting cars out the door. As time went by things got better. Eventually they got to the point where nearly every car coming off the line was perfect.

Around the same time companies like Mercedes Benz were devoting no less than 25% of their factory floor towards fixing manufacturing defects. During this same time period Toyota was devoting less than 5% of their factory floorplan to fixes. This, ironically, meant that the Japanese company was probably delivering a higher quality and more reliable product than the high end car manufacturer.

Source: Read a book on the subject many years ago: "The Machine that changed the world" https://goo.gl/VQw6HS

Back to Linode. The fact that they've seen so many attacks gives them the opportunity to become far better when compared to someone who never sees attacks. Whether they go there or not is entirely in their hands. If this is what's happening they need to take the time to communicate it to the world.

There are at least two possible attitudes: The first is to live in fear of attacks and hope they never happens. The second is to embrace them as an opportunity to get better.

The latter is a formula for disaster, or mediocrity at best. The latter is supported by millions of years of history on this planet showing that things get better because something or someone had to find a solution to a problem.

You can't become a good sailor without facing a few storms.

I'm sticking with Linode.

sb8244• 10 years ago

The former is a recipe for disaster.

rebootthesystem2• 10 years ago

Darn! Too late to edit.

Thanks.

ksec• 10 years ago

It cant be a customers of Linode that is being targetted, Linode has seen WAY MORE DDOS attack then any other web host.

Someone must have hated them or a competitor is trying to destroy them.

These frequent attack just isn't normal.

alexforster• 10 years ago

I appreciate the kind words in many of the comments here, and I'm sure my colleagues do too.

For reference, our third party monitoring measured approximately 23 minutes of downtime–

  1m 46s @ 2016-09-08 01:37:53 EDT
  1m 10s @ 2016-09-08 01:38:14 EDT
  17m 4s @ 2016-09-08 01:40:28 EDT
  2m 11s @ 2016-09-08 01:58:29 EDT

We are now capable of absorbing most volumetric attacks toward our web infrastructure. This outage was a layer 7 attack that needed to be manually mitigated.

vorotato• 10 years ago

I wonder if it has to do with people using linode as a way to get around a particular country's censorship.

Linode Manager and API are under attack

Related Stories