back to article Google cloud TITSUP

Google's cloud has suffered an hour-long outage. The incident started as a connectivity problem with the Alphabet subsidiary's Cloud VPN at 18:51 Monday, Pacific time, in the asia-east1 region. 30 minutes later, Google reported “severe network connectivity issues in all regions.” 20 minutes later, services were restored. …

  1. jake Silver badge

    Cool!

    I need to purchase a heaping helping of that!

    "clouds" must be totally awesome!

  2. RegGuy1 Silver badge

    Hmm...

    Did nobody tell them they needed to buy two of everything?

    1. Adam 52 Silver badge

      Re: Hmm...

      In this case two of everything wouldn't have helped as they lost more than one region. That's bad, very bad. It means anyone looking for reliability can't go with a single supplier if that supplier is Google.

      ...but they did fix it in roughly an hour, and having spent that much time on hold to C&W for a multi link outage before I wonder how most internal IT teams would have coped.

  3. Anonymous Coward
    Anonymous Coward

    The Cloud...

    Other people's computers you have no control over.

    1. I ain't Spartacus Gold badge

      Re: The Cloud...

      That's why I use it. We're a very small company, but we can get a lot of benefits from IT that we're simply not competent to manage ourselves. So our choice is to go with a small local company, and get more cock-ups but more personal service - and we've done that and it was fine. Or a larger company, and get less personal service, but benefit from economies of scale and better uptime, hopefully only one major outage every couple of years.

      IT problems land up on my desk, along with my real job, and I know my level of competence - so just keep the laptops and phones running. Everything else gets outsourced - or we do without. Companies like ours couldn't have mobile accessible CRM 10 years ago, and it's worth it - even with the odd problem. So far one 25 minute outage, since we dumped our own server two years ago. And Office 365 hasn't noticeably let us down once. Yet...

      Virgin killed our server's network access about 5 times in the last year we had it. The last one for 6 hours. If the office network or power goes down, we can just work from home or use 4G. So we're more resilient than having the server onsite.

      Were I running a company large enough to have a decent sized IT department, I'd want the control and ability to manage risk that comes with your own kit. Though even there, I bet there are some things that aren't so critical, and it may be worth some cheap cloud. But you've got to be a pretty big company to be able to afford a backup datacentre with fail-over so that you can theoretically avoid all outages. If you've got the balls to actually test your fail-over system regularly, so that it actually works when needed...

  4. Daggerchild Silver badge
    Flame

    HOUR long? Pfah! Amateurs.

    The Symantec(/Messaglabs) Cloud user control console has been dead for about a DAY now. Database on fire. Users blind. Unicorns screaming, popcorn everywhere!

    https://clients.messagelabs.com/

    1. Andy Neale
      FAIL

      Re: HOUR long? Pfah! Amateurs.

      Symantec are doing a great job of not knowing when it will be back up. ...Which one of their admins opened a cryptolocker email is still open to question.......

  5. Sil

    It apparently affected Discord.

  6. Anonymous South African Coward Bronze badge

    One more reason to have both a NAS and cloud storage for your preciouses data

  7. Kevin McMurtrie Silver badge

    Pooping in your Nest

    Were they using Revolv automation in the datacenter?

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon