MDE Down all morning 4/12/17

[Casper Gutman]

You know, there is a way for Disney to never have another problem with FP+ ever again...

 

[Laketravis]

You know, there is a way for Disney to never have another problem with FP+ ever again...

That would be awesome.

 

[MinnieWaffles]

Oh no, poor people in a theme park who have to WAIT IN LINE because their fastpass is down! Whatever will you DO??

 

[Animaniac93-98]

Hey...on the bright side, I think you have a free pass to just absolutely choke the living crap out of someone. So, you got that going for you...

You do not have that free pass. Please don't choke anyone.

But if he doesn't, will I get a refund on my Fireworks Choking Viewing Dessert Party package?

Too late. I only checked the fine print after I put that jerk weed into a sleeper hold.

Dang, I missed it!

 

[Laketravis]

Actually - 99.99% is a MINIMUM for a commercial e-commerce system, Disney does not even get to 'Two Nines'

View attachment 199469

My bet is if the backend is ISO certified then they are shooting for Six Sigma levels. I can't recall Amazon ever being down and I use it much more often than I use the Disney website. But the frequency of downtime for the Disney site is visually higher. And a 3 minute queue for an automated web process? How does that happen?

 

[GhostHost1000]

Couldn't agree more. This system failed for 5 hours one day in April of 2016. I know, because it was the only day we had in the MK. So this is not an isolated Spring Break incident.

If the reported price tag of $2 billion for this system is correct, a number of consulting companies are laughing all the way to the bank. If you tell your customer that they have to rely on your technology to experience what they have already paid for, then that technology better work correctly 99.9 % of the time. If it doesn't, pro-rated refunds should be made to the registered credit card for the lost time.

Do you think Visa, MasterCard or Discover would stand for a 3 hour outage in prime time? No. There are safeguards and two, sometimes three levels of redundancy built into those systems. These companies due stress testing to make sure their systems do not fail from Thanksgiving to Christmas Eve. One would think Disney would stress test and build to the Spring Break demands.

This has nothing to do with the 2bil project. This is an issue they have across the board with all their websites and even before that

 

[Laketravis]

This has nothing to do with the 2bil project. This is an issue they have across the board with all their websites and even before that

I do recall where website upgrades were included in that project. Would have been an excellent opportunity to address chronic downtime issues.

 

[ford91exploder]

I wonder these people will become frequent flyers. I'm glad they had you in their corner. But what about the other first timers who get the shaft?

Those first timers will not be returning

 

[ford91exploder]

My bet is if the backend is ISO certified then they are shooting for Six Sigma levels. I can't recall Amazon ever being down and I use it much more often than I use the Disney website. But the frequency of downtime for the Disney site is visually higher. And a 3 minute queue for an automated web process? How does that happen?

Most large websites use devices known as load balancers to aggregate multiple physical servers into a single logical server. If there are no available physical resources its possible to place your request into a queue until resources become available.

In short insufficient capacity for the offered load

 

[ford91exploder]

I do recall where website upgrades were included in that project. Would have been an excellent opportunity to address chronic downtime issues.

Remember the apologists keep telling us about how web and infrastructure updates were part of that 3.5 billion boondoggle

 

[ford91exploder]

How many total hours in a year has this happened?



Speaking for myself only.. I do the same thing when I'm using my phone for directions.. I always screen shot the "list view" just in case.

Premium products still have glitches. It's few and far between. It was 3 hours at most? People have extremely high expectations for compensation for a downed app for a few hours. I'm kind of surprised by it.

The OP received "a few" (not sure if that means 3) anytime fast passes, which is pretty much gold on a day like today...and has way more worth than the single 3rd FP that he forgot what attraction it was even for.
I'm not blaming him, but I mean, his fastpass wasn't taken, he just didn't know where to go to use it.
Is that not enough compensation in your opinion?

Based on historic trends Disney park systems are usually down 4-6 days total time for the entire year. This morning guests were lucky as it was just MDE down. Usually it all drops including ADR's and charging

 

[RustySpork]

Actually - 99.99% is a MINIMUM for a commercial e-commerce system, Disney does not even get to 'Two Nines'

View attachment 199469

Not to mention that every 9 is at least an order of magnitude more expensive to achieve and maintain than the previous. Meeting 90% uptime is good enough when you have the fanbase that excuses everything and anything that they do wrong.

They've built a brand so strong that they don't care that they don't have 20ns link failover and zero ping loss between clusters. They probably don't even have basic interface redundancy, let alone clusters or DR capability. Why would they, they don't need it.

 

[DisneyJoe]

Based on historic trends Disney park systems are usually down 4-6 days total time for the entire year. This morning guests were lucky as it was just MDE down. Usually it all drops including ADR's and charging

The travel booking engine is down much more than that.

 

[RustySpork]

How many total hours in a year has this happened?

There have been multiple blips and outages this month alone reported here in the threads and around the intertubes.

 

[Casper Gutman]

There have been multiple blips and outages this month alone reported here in the threads and around the intertubes.

It's ALMOST like it wasn't the best idea in the world to tie the guest experience largely to a web site's reliability

 

[Laketravis]

Most large websites use devices known as load balancers to aggregate multiple physical servers into a single logical server. If there are no available physical resources its possible to place your request into a queue until resources become available.

In short insufficient capacity for the offered load

And many have moved to AWS and enjoy literally no downtime at all.

 

[Bob]

And many have moved to AWS and enjoy literally no downtime at all.

Amazon AWS S3 outage is breaking things for a lot of websites and apps

 

[Laketravis]

Amazon AWS S3 outage is breaking things for a lot of websites and apps

LOL! Yeah, I remember that. It affected quite a few. But S3 isn't the only AWS service available. Probably not a service tier that Disney would use - but then again

 

[larryz]

Hey...on the bright side, I think you have a free pass to just absolutely choke the living crap out of someone. So, you got that going for you...

You do not have that free pass. Please don't choke anyone.

You may have accidentally confused parade watching with buying a ticket on United...

 

[Bob]

LOL! Yeah, I remember that. It affected quite a few. But S3 isn't the only AWS service available. Probably not a service tier that Disney would use - but then again

I can keep listing outages if you'd like.

AWS outage knocks Amazon, Netflix, Tinder and IMDb in MEGA data collapse