> This incident affects: Network Infrastructure (South Korea North (Chuncheon), India South (Hyderabad), Australia Southeast (Melbourne), India West (Mumbai), Japan Central (Osaka), South Korea Central (Seoul), Japan East (Tokyo), Australia East (Sydney), Canada Southeast (Montreal), Canada Southeast (Toronto), Netherlands Northwest (Amsterdam), Germany Central (Frankfurt), Switzerland North (Zurich), UAE East (Dubai), Saudi Arabia West (Jeddah), Chile Central (Santiago), Brazil East (Sao Paulo), UK West (Cardiff), UK South (London), US East (Ashburn), US West (Phoenix), US West (San Jose)) and
> Console (South Korea North (Chuncheon), India South (Hyderabad), Australia Southeast (Melbourne), ap-mumbai-1, Japan Central (Osaka), ap-seoul-1, Australia East (Sydney), ap-tokyo-1, Canada Southeast (Montreal), ca-toronto-1 region, Netherlands Northwest (Amsterdam), eu-frankfurt-1 region, eu-zurich-1 region, UAE East (Dubai), Saudi Arabia West (Jeddah), Chile Central (Santiago), sa-saopaulo-1, UK West (Cardiff), uk-london-1 region, us-ashburn-1 region, us-phoenix-1 region, US West (San Jose)).
So basically the entirety of Oracle Cloud is unreachable. Oops.
Edit: If you were wondering if Oracle really has that many regions, it's actually the same list repeated twice for "Networking" and "Console". Which means both the data and the control plane seem to have fallen off the face of the earth.
When I heard about Oracle Cloud I'll admit I took a quick-look to see if they had any USP for "general" cloud computing customers - or even if they'd be significantly cheaper than our Azure and AWS subs to make the cost to transition worthwhile (lol, of course not)...
The main reason we won't consider using Oracle Cloud is simply because Oracle has already demonstrated their sheer contempt for their own customers through their on-prem database licensing program, so I dread to think what they would do to enforce their IP rights or ToS with their cloud customers.
Another reason we disregarded them (and GCP[1]) is because they simply aren't _big and successful enough_ to give us the confidence they'll keep on running into the forseeable future and with the kind of reliability we expect - and this outage incident demonstrates that they clearly aren't. I know that Azure and AWS have both had significant outage incidents in the past, but both have been rock-solid stable in recent years. For this to happen at Oracle suggests something is institutionally wrong: if they can't get 99.999% uptime on their own worldwide Cloud operations then I cannot trust them with our own SaaS platform because we promise 99.999% to our own customers.
[1]I'm wary about GCP because Google hasn't issued any kind of promissory statement that GCP won't end up on the [Google Graveyard](https://killedbygoogle.com/)
The terms of use for GCP does say any product that is shut down will give at least a one-year notice before it's shut down[1]. For any given product within GCP that has been shut down the lead time has been greater than a year as far as I'm aware.
The fact that Google is calling out GCP separately on its earning calls is actually somewhat of a commitment to trying to grow that business. Very few products within Google are called out separately, so that was a fairly big change.
> (d) Discontinuation of Services. Google will notify Customer at least 12 months before discontinuing any Service (or associated material functionality) unless Google replaces such discontinued Service or functionality with a materially similar Service or functionality.
> Another reason we disregarded them (and GCP[1]) is because they simply aren't _big and successful enough_ to give us the confidence they'll keep on running into the forseeable future
I've got a great big bottle of schadenfreude waiting for the day Goog announce that GCP was just somebody's 20% project and they're moving to another team - complete with an "Our fantastic journey" blog post and a new entry on all the google graveyard websites...
Fwiw, the amount of money that Google is pouring into GCP, including their own transoceanic fiber makes me believe they are committed to it more than “other” endeavors.
Honestly the biggest issue with Google is that they need to dramatically improve their open source database offerings, including 3rd party vendors.
RDS/Aurora eats their proverbial lunch. Azure at least bought Citus to improve their PG offering.
The GCP database offerings feel like something that was put there to check the box, not to be competitive.
> I'm wary about GCP because Google hasn't issued any kind of promissory statement that GCP won't end up on the [Google Graveyard](https://killedbygoogle.com/)
Isn't that literally what an SLA is, a (contractually) guaranteed level of service for some interval of time?
It’s one thing to be wary of experimental new consumer products like many Google has tried and discarded, but I can assure you I have personally observed vast amounts of money poured into GCP in the enterprise realm and even if they retire a GCP product here and there, most of the myriad products in GCP will be around making money for quite a while!
From what I've seen, Oracle Cloud usage is mostly to lift-and-shift legacy Oracle software packages off-prem and things that need low-latency access to them.
It's more Oracle software as SaaS, some basic IaaS (more like a hosting provider than a cloud) and hosted Oracle databases (which are just Oracle running Exadata for you).
So it's big companies struggling to get off old Oracle (and acquisitions like PeopleSoft) stuff and it's gross.
I don't know of anyone using it as "their cloud", but more of another thing to mix in with AWS/Azure/GCP to get rid of on-prem or colo facilities. It's more like IBM Cloud: A way to hang on to customers, sell services (human services like consulting and support), and keep milking those old products for big companies.
We do. Our stack is a fairly standard Java app on an oracle dB. Seems to work well enough. We couldn't get the app to work with azure and aws is a no go (we have retail clients)
I haven't investigated in more detail but Oracle Cloud supposedly has lower bandwidth costs than AWS and Azure, making it cheaper in the long run for transfer heavy services. That's why Zoom uses it.
I'm using it for over one year now as my backup remote programming platfrom and CI server for small projects (Jenkins), and on other server as mysql + minecraft server machine.
Working flawlessly with ssh, out of the box.
It is FAR simpler to manage than AWS in my opinion.
Off-topic, but is there a reason for small companies/projects to use Oracle Cloud? Are their prices and feature set competitive against AWS, Azure and GCP for this audience?
Looking at that page, the total downtime was around 27 minutes. Having been through multiple AWS outages lasting many hours, this is not even a blip on the radar. Except for the usual disparaging of all Oracle products, it is a non story .
Was anyone here impacted by this and could kindly describe what they've experienced?
Was it just actually DNS resolution issues - meaning that stuff working with IP addresses was fine? Or was it just all unreachable and a black hole?
I would imagine DNS plays quite a key role in the internal operations of literally everything in a public cloud but I could be wrong.
Once this actually comes back online, if you're looking for free stuff, Oracle Cloud gives you the most free resources of any cloud provider. But probably you should have a backup.
TIL. I was so excited to try this out, finally a cloud provider with decent pricing (especially considering I'm on South America). Definitely the best price around here by a long margin.
But then I just spent 20 minutes trying to sign up. Every time with my credit card being rejected due to "wrong billing address". Never had this happen before to me, I tried pretty much every permutation of my address. Sigh, I guess Oracle will always be Oracle.
[+] [-] Clewza313|5 years ago|reply
> This incident affects: Network Infrastructure (South Korea North (Chuncheon), India South (Hyderabad), Australia Southeast (Melbourne), India West (Mumbai), Japan Central (Osaka), South Korea Central (Seoul), Japan East (Tokyo), Australia East (Sydney), Canada Southeast (Montreal), Canada Southeast (Toronto), Netherlands Northwest (Amsterdam), Germany Central (Frankfurt), Switzerland North (Zurich), UAE East (Dubai), Saudi Arabia West (Jeddah), Chile Central (Santiago), Brazil East (Sao Paulo), UK West (Cardiff), UK South (London), US East (Ashburn), US West (Phoenix), US West (San Jose)) and
> Console (South Korea North (Chuncheon), India South (Hyderabad), Australia Southeast (Melbourne), ap-mumbai-1, Japan Central (Osaka), ap-seoul-1, Australia East (Sydney), ap-tokyo-1, Canada Southeast (Montreal), ca-toronto-1 region, Netherlands Northwest (Amsterdam), eu-frankfurt-1 region, eu-zurich-1 region, UAE East (Dubai), Saudi Arabia West (Jeddah), Chile Central (Santiago), sa-saopaulo-1, UK West (Cardiff), uk-london-1 region, us-ashburn-1 region, us-phoenix-1 region, US West (San Jose)).
So basically the entirety of Oracle Cloud is unreachable. Oops.
Edit: If you were wondering if Oracle really has that many regions, it's actually the same list repeated twice for "Networking" and "Console". Which means both the data and the control plane seem to have fallen off the face of the earth.
[+] [-] victor9000|5 years ago|reply
[+] [-] sylens|5 years ago|reply
[+] [-] DaiPlusPlus|5 years ago|reply
The main reason we won't consider using Oracle Cloud is simply because Oracle has already demonstrated their sheer contempt for their own customers through their on-prem database licensing program, so I dread to think what they would do to enforce their IP rights or ToS with their cloud customers.
Another reason we disregarded them (and GCP[1]) is because they simply aren't _big and successful enough_ to give us the confidence they'll keep on running into the forseeable future and with the kind of reliability we expect - and this outage incident demonstrates that they clearly aren't. I know that Azure and AWS have both had significant outage incidents in the past, but both have been rock-solid stable in recent years. For this to happen at Oracle suggests something is institutionally wrong: if they can't get 99.999% uptime on their own worldwide Cloud operations then I cannot trust them with our own SaaS platform because we promise 99.999% to our own customers.
[1]I'm wary about GCP because Google hasn't issued any kind of promissory statement that GCP won't end up on the [Google Graveyard](https://killedbygoogle.com/)
[+] [-] kyrra|5 years ago|reply
The terms of use for GCP does say any product that is shut down will give at least a one-year notice before it's shut down[1]. For any given product within GCP that has been shut down the lead time has been greater than a year as far as I'm aware.
The fact that Google is calling out GCP separately on its earning calls is actually somewhat of a commitment to trying to grow that business. Very few products within Google are called out separately, so that was a fairly big change.
[1] https://cloud.google.com/terms/
> (d) Discontinuation of Services. Google will notify Customer at least 12 months before discontinuing any Service (or associated material functionality) unless Google replaces such discontinued Service or functionality with a materially similar Service or functionality.
[+] [-] bigiain|5 years ago|reply
I've got a great big bottle of schadenfreude waiting for the day Goog announce that GCP was just somebody's 20% project and they're moving to another team - complete with an "Our fantastic journey" blog post and a new entry on all the google graveyard websites...
[+] [-] brightball|5 years ago|reply
Honestly the biggest issue with Google is that they need to dramatically improve their open source database offerings, including 3rd party vendors.
RDS/Aurora eats their proverbial lunch. Azure at least bought Citus to improve their PG offering.
The GCP database offerings feel like something that was put there to check the box, not to be competitive.
EDIT: The non-Google database offerings.
[+] [-] jaytaylor|5 years ago|reply
https://www.oracle.com/cloud/free/
Could be worth checking out.
[+] [-] cett|5 years ago|reply
Would you avoid AWS because of Fire Phone?
[+] [-] cmeacham98|5 years ago|reply
Isn't that literally what an SLA is, a (contractually) guaranteed level of service for some interval of time?
[+] [-] stephenhuey|5 years ago|reply
[+] [-] us0r|5 years ago|reply
Snapchat paying them $400 MILLION per year takes care of this issue. That's 1 company. A few jaw dropping others have popped up in S-1 filings.
[+] [-] sjg007|5 years ago|reply
[+] [-] machineko|5 years ago|reply
[+] [-] samurai_noobie|5 years ago|reply
[+] [-] redwood|5 years ago|reply
[+] [-] salmo|5 years ago|reply
It's more Oracle software as SaaS, some basic IaaS (more like a hosting provider than a cloud) and hosted Oracle databases (which are just Oracle running Exadata for you).
So it's big companies struggling to get off old Oracle (and acquisitions like PeopleSoft) stuff and it's gross.
I don't know of anyone using it as "their cloud", but more of another thing to mix in with AWS/Azure/GCP to get rid of on-prem or colo facilities. It's more like IBM Cloud: A way to hang on to customers, sell services (human services like consulting and support), and keep milking those old products for big companies.
[+] [-] geogra4|5 years ago|reply
[+] [-] rospaya|5 years ago|reply
https://www.lastweekinaws.com/blog/why-zoom-chose-oracle-clo...
[+] [-] hansor|5 years ago|reply
Working flawlessly with ssh, out of the box. It is FAR simpler to manage than AWS in my opinion.
There is "no catch".
[+] [-] spyridonas|5 years ago|reply
[+] [-] hermanradtke|5 years ago|reply
[+] [-] nojito|5 years ago|reply
Order of magnitude cheaper and way more predictable costs than other cloud providers.
[+] [-] stjohnswarts|5 years ago|reply
[+] [-] NearAP|5 years ago|reply
Companies also use Oracle Cloud PaaS to build custom enterprise applications
[+] [-] suyash|5 years ago|reply
[+] [-] xbar|5 years ago|reply
[+] [-] 123456789abcde|5 years ago|reply
[+] [-] inetsee|5 years ago|reply
[+] [-] DaiPlusPlus|5 years ago|reply
[+] [-] dgellow|5 years ago|reply
[+] [-] davecheney|5 years ago|reply
[+] [-] haolez|5 years ago|reply
[+] [-] quaffapint|5 years ago|reply
[+] [-] super_linear|5 years ago|reply
[+] [-] ab_testing|5 years ago|reply
[+] [-] coilpancake|5 years ago|reply
[+] [-] imwillofficial|5 years ago|reply
[+] [-] BluePen8|5 years ago|reply
[+] [-] trevorsstar|5 years ago|reply
[+] [-] Aperocky|5 years ago|reply
[+] [-] irrational|5 years ago|reply
[+] [-] Narkov|5 years ago|reply
[+] [-] mainde|5 years ago|reply
I would imagine DNS plays quite a key role in the internal operations of literally everything in a public cloud but I could be wrong.
[+] [-] 0xbadcafebee|5 years ago|reply
[+] [-] suyash|5 years ago|reply
[+] [-] ss1234|5 years ago|reply
[+] [-] napsterbr|5 years ago|reply
But then I just spent 20 minutes trying to sign up. Every time with my credit card being rejected due to "wrong billing address". Never had this happen before to me, I tried pretty much every permutation of my address. Sigh, I guess Oracle will always be Oracle.
[+] [-] Redoubts|5 years ago|reply
[+] [-] linguistbreaker|5 years ago|reply