Meta outage • Hacker News Board

Re: Meta outage

#391

Post by govg » Tue, Mar 05, 2024, 4:22 PM UTC

DeusExMachina wrote: post #281

Looking at the Downdetector home page [1], it looks like many more services are having outages, not just the ones owned by Meta, including: - Google - YouTube - Google Play - T-Mobile - X (Twitter) - Discord - TikTok - Pokemon Go - Snapchat It looks like they all have the same failure point. [1] https://downdetector.com

There is no consistent scale on that graph, so any local maxima of reports received would look similar to any other.

Re: Meta outage

#392

Post by boring_twenties » Tue, Mar 05, 2024, 4:22 PM UTC

bigwheeler wrote: post #334

Tip for Meta engineers: when your service is failing, don't just log people out and prevent logins. Display a cute image that shows that the service is drastically failing (like a whale or something), and then people will know to stop trying to repeatedly log in. The public might even come up with a catchy name for the whale.

Beyond unbelievable that going on an hour later, they're still showing "incorrect password" errors. How many hundreds of millions of people have wasted time frantically trying (in vain) to reset their passwords and pointlessly freaking out that their account might be compromised? What a bunch of careless, incompetent excuses for engineers.

Re: Meta outage

#393

Post by hsuduebc2 » Tue, Mar 05, 2024, 4:22 PM UTC

Well if everything down maybe isn't problem at services itself but at the DNS side?

Re: Meta outage

#394

Post by vb-8448 » Tue, Mar 05, 2024, 4:22 PM UTC

Seems like even the status page doesn't work ... i'd say typical for this type of situations :D :D :D

Re: Meta outage

#395

Post by mlucas82 » Tue, Mar 05, 2024, 4:23 PM UTC

Frankly, I don't buy this explanation - technically or logically both from a devops,and systems architecture level - There is no way in hell a company like Meta is pushing database design changes this significant to production? We all know how many times these database architecture changes get run in staging,then even production subsets before rolling out to production at large.

Should we assume the teams working to ensure Business Continuity & Applications Resiliency redundancies fell asleep at the wheel?

Also, to assume that no down or outage messaging go out during a fairly routine maintenance based outage?

I call BS

Lol I can't find scenario where this happens, at this scale at a company of Meta's scale...

Re: Meta outage

#396

Post by getlawgdon » Tue, Mar 05, 2024, 4:23 PM UTC

Waterluvian wrote: post #286

albatros1313 wrote: post #238

I love how no one here is talking about the "coincidental" timing that this is happening on Super Tuesday. Carry on, nothing to see here.

Same reason we're also not talking about the ten million non-coincidences that didn't seem to happen today. It's a kind of survivorship bias that coincidences seem like they mean something. Or you could imagine it as this: your animal brain has evolved to notice patterns. Seeing coincidences like this is akin to seeing faces in the stars. The challenge of evolved being is to override those impulses when they're not l…

Sure. But if millions of evolved brains are "seeing the same pattern", they will behave as such, and so the meaning of that otherwise random noise in fact becomes something material.

Re: Meta outage

#397

Post by mihaic » Tue, Mar 05, 2024, 4:24 PM UTC

Woshiwuja wrote: post #195

fishnchips wrote: post #176

Earlier quoted context omitted.

Negligible. It would have gone after Twitter first.

You mean X right?

I don't think that anyone but Elon ever means X.

Re: Meta outage

#398

Post by newman8r » Tue, Mar 05, 2024, 4:24 PM UTC

bigwheeler wrote: post #334

Tip for Meta engineers: when your service is failing, don't just log people out and prevent logins. Display a cute image that shows that the service is drastically failing (like a whale or something), and then people will know to stop trying to repeatedly log in. The public might even come up with a catchy name for the whale.

How many other people here just assumed they had been banned for some arbitrary, uncontestable reason? I just use instagram to post hiking photos...

Re: Meta outage

#399

Post by sumitgt » Tue, Mar 05, 2024, 4:24 PM UTC

bigwheeler wrote: post #334

Tip for Meta engineers: when your service is failing, don't just log people out and prevent logins. Display a cute image that shows that the service is drastically failing (like a whale or something), and then people will know to stop trying to repeatedly log in. The public might even come up with a catchy name for the whale.

Could be an outage in the auth service itself

Re: Meta outage

#400

Post by passwordoops » Tue, Mar 05, 2024, 4:24 PM UTC

sbrother wrote: post #385

DeusExMachina wrote: post #281

Looking at the Downdetector home page [1], it looks like many more services are having outages, not just the ones owned by Meta, including: - Google - YouTube - Google Play - T-Mobile - X (Twitter) - Discord - TikTok - Pokemon Go - Snapchat It looks like they all have the same failure point. [1] https://downdetector.com

HN seems to be struggling too, but that could just be everyone here to talk about the outages.

Happens whenever Sama sneezes too