Google, AWS, Cloudflare Among Popular Services Hit By Widespread Outage
- Reference: 0178024323
- News link: https://it.slashdot.org/story/25/06/12/194251/google-aws-cloudflare-among-popular-services-hit-by-widespread-outage
- Source link:
More details to follow.
[1] https://downdetector.com/
Scrapers out of control? (Score:2)
There's been a suspicious amount of scraper traffic the last few days, someone somewhere is running a massive botnet.
Re:Scrapers out of control? YUPPERS!!!! (Score:3)
I'm seeing TONS of it, along with massive coordinated attacks and scans from IP's all over the world.
Any coming from Russian China, North Korea - get a permanent IP block.
Re: (Score:2)
I implemented portsentry feeding fail2ban on edge servers to deal with the unrelenting scans.
It helps, somewhat.
[1]https://github.com/portsentry/... [github.com]
[1] https://github.com/portsentry/portsentry
Search: What day is it (Score:1, Insightful)
Thursday, June 12, 2025 Did we split timelines, again!?
Re: (Score:2)
> Thursday, June 12, 2025
> Did we split timelines, again!?
Nope; just pre-empting tomorrow's internet carnage. Today's mess is something else.
Re: (Score:2)
Depends on where you are. It is the 13th June 2025 here and at the time of your post it was so in many countries. Remember the USA is behind most countries.
To be fair the post does say ET, which I would assume to be the east coast of the USA. Maybe the post was from someone in a different time zone from ET, so accidentally used their local date mixed the time at the location of the issue?
Re: (Score:3)
No. I'm visiting Australia right now, and AEST is only 7 AM right now. When this story was posted, it wasn't 2 PM Friday anywhere in the world.
GCP (Score:1)
Root cause GCP?
[1]https://status.cloud.google.co... [google.com]
[1] https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1SsW
Re: (Score:2)
GCP is definitely being hit by it, but I don't see how it could be the root cause when AWS and Azure are also having problems.
Re: (Score:3)
My theory is that Azure and AWS are having issues due to a large number of multi-cloud GCP customers failing over. They're just overloaded - it's not one or two GCP regions that are down. It's all of them.
Although I've been having issues with AWS all day, since before GCP went down. So I'm not sure.
Re: (Score:2)
Does appear to be just a GCP problem. AWS and Azure's own status pages are showing no problems.
Re: GCP (Score:2)
It's going to be one heck of a post-mortem.
Ah, the modern Internet. (Score:1)
BGP routing updates have taken down huge chunks of the Internet in the past, though I think most major network providers have good BGP update filters in place to prevent a recurrence of the same exact type as has happened a few times in the past.
Nonetheless, it's amazing how much such a distributed, global system can rely on some common underpinnings.
Maybe global over-reliance on common CDN, security platforms, and/or anti-DDOS technology that is being attacked (or affected unintentionally) is causing some
This is fine (Score:2)
The AWS health dashboard shows no issues -- [1]https://health.aws.amazon.com/... [amazon.com]
However my Cloudflare dashboard is completely in accessible and there's a lot of red going on at [2]https://www.cloudflarestatus.c... [cloudflarestatus.com]
But with Cloudflare, that's not unusual... their core DNS and caching services seem unaffected, however.
[1] https://health.aws.amazon.com/health/status
[2] https://www.cloudflarestatus.com/
Internet-wide planned outage (Score:4, Funny)
Each cloud service should have a couple of days per year on which they shut down for 24 hours so everyone can practice their contingency plans.
Re: (Score:1)
source: your ass
Re:Internet-wide planned outage (Score:5, Interesting)
> No, That's an outage that costs money, and can even impact medical devices... which means right now people are dying because of this outage because medical devices aren't able to work correctly.
> Have some respect.
Any organization that places critical medical devices "in the cloud" should be shut down and sued into oblivion. It wasn't too long ago that someone scanned with shodan.io and found IV pumps, PCA pumps, heart monitors, and radiology scanners open to the Internet. Then we just had the story about security cameras in the same regard. Having worked IT in healthcare, I can confirm places that have left all this stuff open, left patient records unencrypted on globally-accessible Windows shares, and worse.
Is this a joke? (Score:4, Informative)
I have Azure services all over the world, no one has complained, and Azure shows no outage or health event. I mean, if no clients complained, how widespread could it be? Maybe it was very brief, so we all missed it, BUT, it would explain why ProtonMail disconnected my external account around 2:30pm EDT.
Re: (Score:2)
Based on what we're seeing, I'm becoming convinced that this is a GCP/Cloudflare problem, and the AWS and Azure reports are just panic.
Re: (Score:1)
I do have issues on AWS us-east, accesing some services, instances and so on. Also, NPM says they ARE implementing a fix (NPM registry is fully dead) [1]https://status.npmjs.org/ [npmjs.org]
[1] https://status.npmjs.org/
Re: (Score:2)
Well, the summary does say it is happening FRIDAY, so stay tuned!
Re: (Score:2)
Ya, I noticed that, but figured it was a typo.
Cloudflare update (Score:3)
Update - Cloudflare’s critical Workers KV service went offline due to an outage of a 3rd party service that is a key dependency. As a result, certain Cloudflare products that rely on KV service to store and disseminate information are unavailable including:
It's only CloudFlare and GCP (Score:1)
AWS was not showing issues at the time all these others were on downdetector.com. None of the 3 AWS regions I use in the US have had issues and there are no notifications in the AWS support center, either. So update your headline. AWS is not involved. It's a CloudFlare and GCP issue. Maybe not Azure either. Those could be CloudFlare issues looking like Azure issues.
Clarification (Score:2)
So this is happening tomorrow?
> began around 2:25pm ET Friday
Oh noes! (Score:2)
Clowns! Clowns striking again!
Why all of them? Seems Suspicious. (Score:1, Funny)
As a network engineer it's a very suspicious to me that every single one of these is having problems at the exact same time, because these are supposed to be on different networks as part of a distributed system that is resilient to a centralized failure.
I think the NSA just admitted to everybody that there's a thing connected to all of them that is looking at all their traffic.
Re: (Score:2, Interesting)
Cloudflare is involved, which could affect a lot of things.
Re: (Score:2)
> Cloudflare is involved, which could affect a lot of things.
Yes, like the governing authority's MITM services are likely offline.
Re:Why all of them? Seems Suspicious. (Score:5, Informative)
It's not all of them. One thing that tends to happen with a major outage is everybody treats downdetector.com as the gospel truth. Then end users try to make their own reports, and don't even know what service they're trying to connect to (like reporting Netflix is down when they try to load Hulu). Then lazy news services just pick up downdetector results and publish stories, which are then picked up and copied by other news services.
What's known: GCP (Google Cloud Platform, not to be confused with [1]Google Actual Cloud Platform [youtu.be]) has experienced a major outage. It affected some Google services as well as a lot of customer services. CloudFlare had a significant outage at the same time, although it looks like mostly just that CF's systems got backed up and overloaded trying to talk to the GCP-hosted customers.
[1] https://youtu.be/Cp10_PygJ4o?feature=shared
Re: (Score:3)
It's also happening tomorrow, according to the summary. Hmmmmmmm...
Re: (Score:2)
Being tomorrow could be just a timezone issue. As I write this I'm having breakfast as it is Friday morning here now, but still Thursday afternoon in the USA.
Re: (Score:3)
> 2:25pm ET Friday
They specified the timezone.
News from the future !