Lots of people getting pop-ups to sign into their account. Seems Apple
servers were down (still are?) last night/early this morning (which
prompted the pop-ups).

--
Cheers … Mark

Mark <captain.black@gmail.com> wrote:
> Lots of people getting pop-ups to sign into their account. Seems Apple
> servers were down (still are?) last night/early this morning (which
> prompted the pop-ups).

Seems OK now as I’ve just logged in there.

<https://www.apple.com/uk/support/systemstatus/>

--
Cheers, Alan

Mark wrote:
> Lots of people getting pop-ups to sign into their account. Seems Apple
> servers were down (still are?) last night/early this morning (which
> prompted the pop-ups).
>

If I were designing a system for a business - even a small one with
fewer than 100 employees - I would put resilience high up my list of
priorities, so that the failure of any one component would not render
the system unavailable to users.

Given that Apple is serving millions of users on an international basis,
one would have thought that resilience was very important, and that
users should never see that the servers are down.

--
Graham J

Graham J <nobody@nowhere.co.uk> wrote:
> Mark wrote:
>> Lots of people getting pop-ups to sign into their account. Seems Apple
>> servers were down (still are?) last night/early this morning (which
>> prompted the pop-ups).
>>
>
> If I were designing a system for a business - even a small one with
> fewer than 100 employees - I would put resilience high up my list of
> priorities, so that the failure of any one component would not render
> the system unavailable to users.
>
> Given that Apple is serving millions of users on an international basis,
> one would have thought that resilience was very important, and that
> users should never see that the servers are down.

It might have been some maintenance work but if so surely it would have
been announced. But you’re correct resilience needs to be way up there.
Apparently the aioe.org news server suffered a major catastrophe some weeks
ago and it’s still not back in service. Maybe it never will?

--
Cheers, Alan

Mark <captain.black@gmail.com> wrote:
> Lots of people getting pop-ups to sign into their account. Seems Apple
> servers were down (still are?) last night/early this morning (which
> prompted the pop-ups).

I got this too in my iPhone late last night.
--
"I will exalt you, my God the King; I will praise your name for ever and ever. Every day I will praise you and extol your name for ever and ever." --Psalm 145:1-2. Is spring break & winter over yet? :) (Orthodox) Easter again! Beat Grizzlies, Lakers!
Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
/\___/\ Ant(Dude) @ http://aqfl.net & http://antfarm.home.dhs.org.
/ /\ /\ \ Please nuke ANT if replying by e-mail.
| |o o| |
\ _ /
( )

Alan B <alanrichardbarker@gmail.com.invalid> wrote:
> Graham J <nobody@nowhere.co.uk> wrote:
> > Mark wrote:
> >> Lots of people getting pop-ups to sign into their account. Seems Apple
> >> servers were down (still are?) last night/early this morning (which
> >> prompted the pop-ups).
> >>
> >
> > If I were designing a system for a business - even a small one with
> > fewer than 100 employees - I would put resilience high up my list of
> > priorities, so that the failure of any one component would not render
> > the system unavailable to users.
> >
> > Given that Apple is serving millions of users on an international basis,
> > one would have thought that resilience was very important, and that
> > users should never see that the servers are down.

> It might have been some maintenance work but if so surely it would have
> been announced. But you???re correct resilience needs to be way up there.
> Apparently the aioe.org news server suffered a major catastrophe some weeks
> ago and it???s still not back in service. Maybe it never will?

More like months ago.
--
"I will exalt you, my God the King; I will praise your name for ever and ever. Every day I will praise you and extol your name for ever and ever." --Psalm 145:1-2. Is spring break & winter over yet? :) (Orthodox) Easter again! Beat Grizzlies, Lakers!
Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
/\___/\ Ant(Dude) @ http://aqfl.net & http://antfarm.home.dhs.org.
/ /\ /\ \ Please nuke ANT if replying by e-mail.
| |o o| |
\ _ /
( )

Graham J <nobody@nowhere.co.uk> wrote:
> Mark wrote:
> > Lots of people getting pop-ups to sign into their account. Seems Apple
> > servers were down (still are?) last night/early this morning (which
> > prompted the pop-ups).
> >
>
> If I were designing a system for a business - even a small one with
> fewer than 100 employees - I would put resilience high up my list of
> priorities, so that the failure of any one component would not render
> the system unavailable to users.

'It's more complicated than that'

Usually systems designed by big tech platforms are hugely complicated, and
designed to be fault tolerant because, at their scale, systems do fail on a
regular basis. There are multiple layers of fault tolerance to handle the
everyday failures, as well as bigger events like a datacentre being on fire.

Resilience is absolutely top of their priorities, that's why there's an
entire class of employees called 'site reliability engineers'.

> Given that Apple is serving millions of users on an international basis,
> one would have thought that resilience was very important, and that
> users should never see that the servers are down.

They do all that. But usually the cause of outages is either a failure mode
nobody thought of, a cascade of things going wrong, or a system not
completely going dark but misbehaving just enough to cause errant behaviour
elsewhere.

One of the bigger points of failure is DNS, because *the entire
internet* relies on DNS working. No DNS: it doesn't matter how fault
tolerant you are if you can't find anyone's IP address. Even if you don't
need DNS to work, every service you're talking to does.

Computers are complicated. Ever more layers of redundancy doesn't make it
less so.

Theo

In article <u1gkcj$2h2q1$1@dont-email.me>, Graham J
<nobody@nowhere.co.uk> wrote:

> If I were designing a system for a business - even a small one with
> fewer than 100 employees - I would put resilience high up my list of
> priorities, so that the failure of any one component would not render
> the system unavailable to users.
>
> Given that Apple is serving millions of users on an international basis,
> one would have thought that resilience was very important, and that
> users should never see that the servers are down.

they definitely do that, but nothing in this world is perfect.

even google has had the occasional outage.

On 16 Apr 2023 at 14:29:06 BST, "Theo" <theom+news@chiark.greenend.org.uk>
wrote:

> Graham J <nobody@nowhere.co.uk> wrote:
>> Mark wrote:
>>> Lots of people getting pop-ups to sign into their account. Seems Apple
>>> servers were down (still are?) last night/early this morning (which
>>> prompted the pop-ups).
>>>
>>
>> If I were designing a system for a business - even a small one with
>> fewer than 100 employees - I would put resilience high up my list of
>> priorities, so that the failure of any one component would not render
>> the system unavailable to users.
>
> 'It's more complicated than that'
>
> Usually systems designed by big tech platforms are hugely complicated, and
> designed to be fault tolerant because, at their scale, systems do fail on a
> regular basis. There are multiple layers of fault tolerance to handle the
> everyday failures, as well as bigger events like a datacentre being on fire.
>
> Resilience is absolutely top of their priorities, that's why there's an
> entire class of employees called 'site reliability engineers'.
>
>> Given that Apple is serving millions of users on an international basis,
>> one would have thought that resilience was very important, and that
>> users should never see that the servers are down.
>
> They do all that. But usually the cause of outages is either a failure mode
> nobody thought of, a cascade of things going wrong, or a system not
> completely going dark but misbehaving just enough to cause errant behaviour
> elsewhere.
>
> One of the bigger points of failure is DNS, because *the entire
> internet* relies on DNS working. No DNS: it doesn't matter how fault
> tolerant you are if you can't find anyone's IP address. Even if you don't
> need DNS to work, every service you're talking to does.
>
> Computers are complicated. Ever more layers of redundancy doesn't make it
> less so.

On 9/11, we had some kit at 25 Broadway, not far from the WTC, where our
x-atlantic circuit terminated. Our onward connection to another provider was
via the Deutche Telekom fibre ring in New York. Too bad both primary and
backup rings of that went via the WTC basement. I could monitor all those
connections from the Cambridge office and watched it all go down, section by
section. At 25 Broadway they switched to the diesel gennies for power, but
after some time (a few hours) these failed and it was discovered that the air
filters had become blockd by dust from the WTC wreckage. Gennie power was then
restored, but again failed after 3 days of so when they ran out of fuel. It
took a little while to get more fuel, which could only come in with a military
escort.

One encouraging aspect during that affair was that those who had working
circuits and fibre in NY were offering them for temporary free use to everyone
else until those with destroyed infrastructure got back on their feet.

--
Tim

Graham J <nobody@nowhere.co.uk> wrote:
> Mark wrote:
>> Lots of people getting pop-ups to sign into their account. Seems Apple
>> servers were down (still are?) last night/early this morning (which
>> prompted the pop-ups).
>>
>
> If I were designing a system for a business - even a small one with
> fewer than 100 employees - I would put resilience high up my list of
> priorities, so that the failure of any one component would not render
> the system unavailable to users.
>
> Given that Apple is serving millions of users on an international basis,
> one would have thought that resilience was very important, and that
> users should never see that the servers are down.

You're very welcome to make a resilient system. If you can do a better job
than Apple you'll have a very long list of customers with deep pockets.