[gmisra]

FYI to S3 customers, per the SLA, most of us are eligible for a 10% credit for this billing period. But the burden is on the customer to provide incident logs and file a support ticket requesting said credit (it must be really challenging to programmatically identify outage coverage across customers /s)

https://aws.amazon.com/s3/sla/

[primitivesuave]

The 10% savings of ~$10 does not compare to time/potential business lost, but thanks for the tip :)

[mabbo]

> potential business lost

My startup's op team had a great discussion today because of this that basically boils down to "if we hit our sales goals, an incident like this a year from now would end our company".

Looks like our plans to start prepping for multi-cloud support will be a higher priority.

[omni]

> an incident like this a year from now would end our company

I'm genuinely curious, what kind of business are you in that a four hour outage would end the company? High frequency trading or something?

[mabbo]

You're in the right ball park- services for traders and brokers in the finance industry. A two hour outage during trading hours would be an extinction level event.

[itake]

not op, but when i worked in the ticketing business if this happened during a big on sale we would get hosed.

[kevan]

A low-cost first step is enabling cross-region replication [1].

[1] http://docs.aws.amazon.com/AmazonS3/latest/dev/crr.html

[owenmarshall]

I'd be willing to bet that the effort you put into a multi-cloud solution will be more expensive than you think, and far more brittle in the event of an emergency. It always is.

[mabbo]

You're not wrong, but if your customers can't afford an outage, you can't afford not having a fallback plan. Probably worth a few trial runs during low volume hours to be sure.

[machbio]

thats for below 99.9% - they are at 99.997% .. you are never getting that 10% credit..

[christop]

0.1% of 28 days is 40 minutes, so it seems likely to happen.

[machbio]

I was calculating it for a year - maybe the availability applies to per billing cycle - you may be correct..

[joatmon-snoo]

You got your orders of magnitude wrong ;)

    99.9964583 = 100 - 153/(30*24*60)
    99.6458333 = 100 * (1 - 153/(30*24*60))

[machbio]

I was calculating it for a year - maybe the availability applies to per billing cycle.. I am still not able to understand your math - mind explaining

[joatmon-snoo]

Your numbers are still off for a year.

    99.9997089 = 100 - 153/(365*24*60)
    99.9708904 = 100 - 100*153/(365*24*60)

The formula is 100% minus 100% times downtime/time in month/year.

153 is the number of minutes they were down going off the reported updates at https://status.aws.amazon.com/ - 11:35AM PST was when they fixed the status page, 2:08PM PST was when S3 was fully back online. (And 153 is underestimating it, because there were errors going on for long before they fixed the status page, but I don't have timestamps on that.)