You are signed in. Close

You are signed out. Close

Thank you. Your subscription has started. Close

Thank you. You will get an email when a new comic comes out. Close

We could not confirm your payment. If you paid, use the link "Sign in" at the bottom of the page. Close

Something went wrong on our side. Please try again in a moment. Close

Previous episode DNS failure: how one empty record stopped a cloud region

Why one cloud outage takes down half the Internet

Race condition: why three copies of one program did not prevent the AWS outage

AWS ran three copies of the program behind one of its most important DNS records, to be safe. How did copies kept for safety end up emptying that record?

  1. 1 Behind the name of DynamoDB stand two programs of AWS. The first, the Planner, keeps an eye on the machines of DynamoDB and writes a fresh plan every so often: the list of addresses the name should give.the Plannera plana plana new plan, again and again

    Behind the name of DynamoDB stand two programs of AWS. The first, the Planner, keeps an eye on the machines of DynamoDB and writes a fresh plan every so often: the list of addresses the name should give.

  2. 2 The second, the Enactor, copies the newest plan into the DNS record, over the one before. To be safe, AWS ran three copies of it, one in each of three zones: if one stops, the two others go on.zone Azone Bzone Cthree copies of the Enactordynamodb.us-east-1.amazonaws.comthe addresses of the newest plan

    The second, the Enactor, copies the newest plan into the DNS record, over the one before. To be safe, AWS ran three copies of it, one in each of three zones: if one stops, the two others go on.

The rest of this episode is for subscribers

Subscribe to read every premium episode. You can cancel when you want.

Paid yearly: $99 instead of $119.88

Next episode Cascading failure: why a 3-hour DNS problem became a 14-hour cloud outage

Share this comic

More comics