Live data from Hacker News

Google outage – resolved

news.ycombinator.com

421–430 of 870 posts

Re: Google outage – resolved

#421

So, anybody still feel like arguing that 'the cloud' is a viable back-up? Or is that a sore point right now? Just for a moment imagine: what if it never comes back again? Of course it will, - at least, it better - but what if it doesn't? And if it does, are you going to take countermeasures in case it happens again or is it just going to be 'back to normal' again?

I guess a lot of people are fine with the risk.

Everybody uses it, so if, like, Gmail loses all the emails, we are then in such a state that the consequences will be more bearable and socially normal.

Most people are fine with accepting that whatever future thing will happen to most people will also happen to them. Because then the consequences will also be normal.

If the apocalypse comes, it comes for almost all of us and that's consolation enough.

Re: Google outage – resolved

#422
This puts into sharp relief how outages and downtime can hit ANYONE.

Cheers to all the small SaaS businesses out there keeping their services up and running without much of a hick-up all year round.

Re: Google outage – resolved

#423
Gcloud/GKE command line utils are also having problems.

Was deploying some test applications and kubectl started complaining about gcloud auth helper throwing not zero errors. Trying to launch cloud shell from the website and nothing happened.

The web application which is a site that does not rely on any external API is running fine.

Re: Google outage – resolved

#424

So, anybody still feel like arguing that 'the cloud' is a viable back-up? Or is that a sore point right now? Just for a moment imagine: what if it never comes back again? Of course it will, - at least, it better - but what if it doesn't? And if it does, are you going to take countermeasures in case it happens again or is it just going to be 'back to normal' again?

What worries me the most is email. I basically don't use any other Google services other than Gmail and YouTube, but for email I really don't know of an alternative.

Sure you can argue "move to Fastmail/Protonmail/Hey/whatever", but those can also go down on you just like Google is down now. And self hosting email is apparently not a thing due to complexity and having to forever fight with being marked as spam (ndr.: not my personal experience, I never tried self hosting, just relying what I read here on HN when the topic comes up).

So, yeah, what do we do about email? I feel like we should have a solution to this by now, but somehow we don't.

Re: Google outage – resolved

#426
Atlassian's #HugOps page[1] collates messages of support for Devops as a public embedded Google Map and it really encapsulates the current feeling!

https://imgur.com/a/3RurTYf

Don't forget to #HugOps all the people who've been woken up on a monday morning with this! Hope this gets resolved soon :)

[1] https://www.atlassian.com/software/statuspage/hugops

Re: Google outage – resolved

#427
post #348

Google is also rejecting SMTP ingress traffic. (delivery temporarily suspended: lost connection with aspmx.l.google.com[2a00:1450:400c:c04::1b] while sending DATA command)

That is Really Bad™. People are going to miss critical emails with no recovery.

These are temporary deferrals, a properly configured outbound mailserver should retry for days.

Re: Google outage – resolved

#429
post #188

Earlier quoted context omitted.

Seems like everyone going on HN now. The site is so slow

Interesting cascade effect of sites going down! I wonder where people will go next to check if HN is down? Also wondering if this is perhaps the fastest upvoted HN post ever? 8 mins -> ~350 votes, 15 mins -> ~750 votes. I wonder if @dang could chime in with some stats? Update: looks like it hit 1000 upvotes in ~25 mins! Update: 1500 in ~40 mins Update: 2000 in ~1 hour 20 mins (used the HN API for the timestamp)

/. , of course.
Post reply on HN