Downdetector spiking hard overnight

https://www.the-independent.com/tech/aws-amazon-internet-outage-latest-updates-b2848345.html

Lots of big names down at once (a whole page), what's happening?

478 points · 60 comments · view on lemmy.world

60 Comments

cabbage@piefed.social · 253 pts · 296d (11 replies)

Seems like a fair bet that Amazon Web Services is having some problems?

I really appreciate that the commercial internet runs on like three service providers. Would love to see it crash and burn.

troed@fedia.io · 107 pts · 296d (9 replies)

Meredith just tooted that Signal is down due to an AWS outage.

aeronmelon@lemmy.world · 111 pts · 296d (2 replies)

Travel back in time 20 years and that sentence makes you sound like a crazy person.

skisnow@lemmy.ca · 51 pts · 296d (1 reply)

Even now it mostly sounds like the sort of line someone in a Black Mirror episode would say

lena@gregtech.eu · 1 pts · 295d

I sometimes catch myself saying programmer-nerdy stuff past me would be incredibly confused about

TheBat@lemmy.world · 16 pts · 296d (2 replies)
[ removed ]
JustJack23@slrpnk.net · 11 pts · 296d

It's more like a dependency of a dependency doesn't and that breaks all relying systems.

troed@fedia.io · 9 pts · 296d

The biggest ad Matrix/Element has ever gotten.

SaharaMaleikuhm@feddit.org · 11 pts · 296d

Toot toot! I still think it's hilarious that this what they are called on mastodon. Great choice

VieuxQueb@lemmy.ca · 8 pts · 295d (1 reply)

Signal uses AWS, fuck ! That's deceiving, Where is security if it runs on Besos hardware ?

MonkCanatella@sh.itjust.works · 16 pts · 295d

Unless they broke encryption as we know it, it's fine. Ethically/morally who knows, is it more important to have a reliable e2ee messaging tool constantly? Probably

CallMeAnAI@lemmy.world · 3 pts · 296d

It was much better with rack space leading the way. We never had outages. Ever.

BozeKnoflook@lemmy.world · 77 pts · 296d (2 replies)

https://health.aws.amazon.com/health/status

I suspect the big problem is that IAM is affected and it is not decentralized, which is causing other systems worldwide (even outside of AWS' us-east-1 location) which rely on IAM in us-east-1 to also fail. I'm having trouble even logging into the AWS console to check on my European servers.

Edit: IAM is the main authentication method. So AWS may still be up and running fine in other locations around the world; but if you can't connect to them because AWS' internal authentication is all fucked up...

JustJack23@slrpnk.net · 10 pts · 296d

Sounds reasonable, the first errors I encountered were all 403.

corsicanguppy@lemmy.ca · 4 pts · 295d

Butbutbut AWS reliability and many regions and zones means this can't happen! ;-)

homesweethomeMrL@lemmy.world · 76 pts · 296d (1 reply)

Me: *watching random thing error out*

Me: *restarting everything, feeling annoyed*

Me: Rats. Well. I guess I’ll see what Lemmy’s got on

Me: Ooohhhhhh

corsicanguppy@lemmy.ca · 4 pts · 295d

I feel this in my midnight cursing and investigation.

I gotta get HAss going so I can still use my lights when the clowd breaks.

JustJack23@slrpnk.net · 56 pts · 296d (3 replies)

Aws us-east-1 seems to have problem, and we can now see what is hosted there.

KyuubiNoKitsune@lemmy.blahaj.zone · 34 pts · 296d (2 replies)

The most used region in AWS, so not surprising

zwerg@feddit.org · 18 pts · 296d (1 reply)

Also the region the PoC their new tech, so its known to be unreliable.

RaivoKulli@sopuli.xyz · 17 pts · 296d

I knew this was due to DEI!

Deebster@infosec.pub · 32 pts · 296d (3 replies)

overnight

Ah, you mean just now. It's not night everywhere!

tym@lemmy.world · 11 pts · 296d (2 replies)
[ removed ]
RaivoKulli@sopuli.xyz · 11 pts · 296d

Oh you mean UTC?

homesweethomeMrL@lemmy.world · 1 pts · 296d

Ach. Fine.

otter@lemmy.ca · 26 pts · 296d (6 replies)

Is Signal down for anyone else?

Reddit seems to be having issues as well. hah

cabbage@piefed.social · 13 pts · 296d

Yup, Signal is down. The one centralized service I'm still rooting for I guess. Disappointed they're running on AWS.

tja@sh.itjust.works · 12 pts · 296d
J4g2F@lemmy.ml · 6 pts · 296d

Yeah same no signal here on signal.

SrMono@feddit.org · 4 pts · 296d

Yea. It‘s affected, too.

SaharaMaleikuhm@feddit.org · 4 pts · 296d (1 reply)

Yup, it's maybe working again now. At least I got a message through just then.

otter@lemmy.ca · 1 pts · 296d

I got the test messages I exchanged with a friend earlier, but new messages are still stuck

General_Effort@lemmy.world · 24 pts · 296d (5 replies)

Outage is ending.

Oct 20 2:27 AM PDT We are seeing significant signs of recovery. Most requests should now be succeeding. We continue to work through a backlog of queued requests. We will continue to provide additional information.

https://health.aws.amazon.com/health/status

JesusTheCarpenter@feddit.uk · 11 pts · 296d (4 replies)

I've just been on Reddit and it's having some serious issues right now.

KaChilde@sh.itjust.works · 27 pts · 296d

Reddit has had issues for years, that’s why so many of us are here.

Oh, you meant the aws thing.

Flatfire@lemmy.ca · 5 pts · 296d (2 replies)

Using the old interface seems to yield better results there. It appears to be their newer API model that's suffering.

Scrollone@feddit.it · 11 pts · 296d (1 reply)

The old Reddit interface is always the best one.

UniversalBasicJustice@lemmy.dbzer0.com · 5 pts · 296d

At this point I'd nearly appreciate the removal of old.reddit or RES. My desktop browsing experience hasn't changed in roughly a decade and a half and that's a shitload of inertia. A few trims of my subscribed list has limited my exposure to the worst of the slop and botting.

Mobile is absolutely a different situation for both Reddit and Lemmy alike. I found Jerboa shortly after the API disaster and have stuck with it since, but I probably need to look into tweaks for desktop/browser Lemmy. RES and old.reddit are a killer combo but fuck their 'official' app and the 'new' interface.

Regardless, I no longer participate on Reddit and highly prefer interacting here; more interesting and informative conversations by far.

mrgoosmoos@lemmy.ca · 18 pts · 295d

US government testing their ability to shut down the internet

Mwa@thelemmy.club · 15 pts · 296d (2 replies)

I assumed Cloudflare tbh

Joeffect@lemmy.world · 19 pts · 296d (1 reply)

But amazon web service was shown down..

https://health.aws.amazon.com/health/status

Having issues

Mwa@thelemmy.club · 2 pts · 296d

Oh

Trying2KnowMyself@lemmy.blahaj.zone · 11 pts · 296d (1 reply)

crabrave

Enzyoo@lemmy.dbzer0.com · 3 pts · 296d

crabrave

dumplingry@lemmy.world · 10 pts · 296d

sad to see my favorite game clash royale down

tal@lemmy.today · 9 pts · 296d
rozodru@piefed.social · 6 pts · 296d

someone likely borked IAM.

abfarid@startrek.website · 5 pts · 295d (2 replies)

Wtf is "whatnot"? How is it amongst such well-known stuff (except Instructure, too, I guess)?

Quexotic@infosec.pub · 1 pts · 295d (1 reply)

With so many eggs being in one basket it really makes me wonder when AWS goes down if it isn't actually something that's an attack when they say something like "oh it's just DNS" because why would they admit it?

tgxn@lemmy.tgxn.net · 3 pts · 295d

Because it was just DNS. That's how the issue presented and that's what it was.

MTZ@lemmy.world · 5 pts · 296d (5 replies)

How long ago did you take the screenshot? Because I am signed in to two of those services right now and they are working 100% fine.

butterycroissant@lemmy.world · 10 pts · 296d (1 reply)

about 10 mins ago

Vinny_93@lemmy.world · 15 pts · 296d

Could it be that web services are divided into regions

Chozo@fedia.io · 5 pts · 296d (1 reply)

A lot of the affected services seem to be partially functional right now. It looks like the AWS network that went offline is one managing real-time analytics data, mostly.

MTZ@lemmy.world · 2 pts · 296d

Ahhhhhhh, I see. That makes sense!

General_Effort@lemmy.world · 4 pts · 296d

Could be that your region relies on different servers, could be that only the sign-in is down.

neon_nova@lemmy.dbzer0.com · 5 pts · 296d (1 reply)

Damn! I was getting pissed at my isp because so much is loading slowly.

I guess this explains the slow loading times for various things.

jjlinux@lemmy.zip · 2 pts · 295d

I was about to restart my whole network while watching something on Stremio. One look at Lemmy was enough to turn that off and pick up a book. Funny how books never buffer 😂

Kolanaki@pawb.social · 5 pts · 296d

I thought VRChat was down, but I can log in on the Quest stand alone version. For some reason, the PC version would flash the login screen then just sit in that blue void not actually loading anything. :/

I also could not connect to Ubisoft to play Bridge Crew and trying to log into UbiConnect won't send the 2FA code to actually get the fuck in.

MonkderVierte@lemmy.zip · 3 pts · 296d

What the? My local internet provider uses Amazon Authentication for their webpage?

spacelord@sh.itjust.works · 3 pts · 296d

Perplexity is down as well.

pewgar_seemsimandroid@lemmy.blahaj.zone · 2 pts · 296d

some people blame [noun] a [verb] games on roblox