Gee, I'm not having any trouble with Libre Office at all... I wonder what the issue is? ;)

" Affected users reported internal mail flowing slowly, while external email had packed up altogether. One wailed: "You got to be kidding me! We haven't gotten emails since 1:30 pm and we run a financial company with clients!!" "

Then what the fuck are you doing putting all of your eggs in to one basket and not taking responsibility for your own communications?

If email is that important to you, you have to take ownership of it and make sure your communications are resilient. Bitching that Office364 isn't working for you yet it's so critical and crucial to your operations just shows that you might not be as diligent with things as you claim to be with your clients.

On the other hand, what kind of financial company doesn't have clients?

Have you shot the board member that insisted that email was a perfectly fine substitue for an API?

A lot of their clients wouldn't have noticed as they'd have been having the same problem.

No one ever got fired for buying Microsoft.

Only because they never got the email....

...because Thunderbird was just showing the message: "Cannot authenticate user".

Then I checked Downdetector.ca... Ah, that explained it.

I guess my next "Weaning myself off Micro$oft" will be to dump Outlook.

Protonmail anyone?

Find a company that can act as registrar, preferably one that can host a mail service for you, get yourself your own domain and have your own mail service, either run by the registrar or, failing that, by another company. If you have problems with your service provider you can simply move to a new provider, keeping your email addresses.

Be picky though, as many providers are just reselling MS-365 MS-364 MS-363 these days.

Pro version of Protonmail allows using your own domain IIRC, for example.

Posteo and Tuta may also be worthy of consideration.

I used Tuta for a year and it was problematic as it forces you to use their own client. It won't work with the iOS client (or any other client regardless).

Find a company that can act as registrar, preferably one that can host a mail service for you, get yourself your own domain and have your own mail service, either run by the registrar or, failing that, by another company.

Yep. Sometimes following the wisdom of the ancients is the right idea. So company needs to understand what services really are business critical. Like an ability to communicate with the outside world. Then understand that there is a cost to managing this risk. This may be higher than the cost of an Outage365 subscription, but that's the delta between cost and risk. But the cost need not be that high to do things the old fashioned way.

It just needs someone who needs to know what a domain name and MX record is, how SMTP works and how to glue all that together. So basically an IT person (or consultant(cy) that can wrangle something other than Microsoft. Or can provide some consultancy to help keep MS stuff from falling over, which it is going to do, so why are so many businesses still using it? Other solutions are available that greatly reduce vendor lock-ins, price gouging for licence fees (one enables the other, naturally) and helps avoid awkward questions from insurers, lenders & shareholders about how the business is managing risk, and why aren't they responding to emails?

(Which is also something I'm kinda hoping will be fallout from the current US v Europe trade spat. Like more governments and businesses deciding they don't want the risks of their MS, AWS or Gmail not working because they've been turned off in a trade/sanctions war.. Which they could easily be, if not for the pressure the US tech titans would apply to ensure that doesn't happen. Then again, it already has. So Russia & Iran can't rent MS, AWS etc licences, so they just use something else instead. Necessity is the mother of invention, sanctions provide that necessity and software development isn't a uniquely Western skillset. See also DPRK and NorkOS(?), which derived from an *nix distro and because hardware is also sanctioned, doesn't need the latest CPU/GPU/AIPU just to display animated emojis in emails, or reasons to tell Copilot to fsck off. But then Copilot seems to be helping wean some corporates off their MS dependency. They don't really want their IP being used as MS 'training data'.)

"It just needs someone who needs to know what a domain name and MX record is, how SMTP works and how to glue all that together."

If you have a decent registrar/MSP supplier there's little of that except setting up mail boxes and addresses when the service is in place. But otherwise, yes. It just nees someone to know that (a) it's possible, (b) it's a really good idea and (c) if you're in business an email address in your company's name looks so much better.

Buy the domain from a seperate vendor from hosting so you can move more easily if either gets bought out by VCs.

No need. They're both portable. It's relying on ISP email that's the real killer, even more than relying of a big US business, but that's because it then makes it really hard to jump ship from the ISP service.

I've had great results with Hover, I wanted to try self-hosting my email, but just couldn't be bothered when they make it so affordable, and so easy. They've been my primary email provider for years, and I get a sick address that no one thinks is nearly as cool as me.

Where's the "pointing and laughing" (AKA Nelson Muntz) icon when you need it?

Its in the mail!

My 2 busiest clients, both of whom use Microsoft359 or whatever the current service level is, have been very quiet today... Guess now I know why.

Really, really hope that an MS outage wasn't the reason I kept getting 404 for the forum pages on all the new Register stories earlier this morning!

Nah, must have been a simple "cleaner pulled the plug" scenario, that is more El Reg worthy.

I've seen that many times before; it appears that the comment page for a particular story isn't immediately created, so early comers get a 404. Give it a few minutes and it'll materialise.

Just like there was/is Castlemaine XXXX (Four X) ...

Microsoft needs to rebrand 'Microsoft 365' as Microsoft XXX (Three X ... 3 followed by a 2 digit number [3??] !!! ) ... this allows for the 365 being in reality 300'ish ... hopefully they can manage to hit 300 days+ as a standard !!!

:)

Lets hope they manage that, not that they need to rebrand even further to Office 2XX-ish

It's secretly Microsoft 3.65 under the hood. And I don't see anybody denying it.

I've been saying for years that migrating away from self-hosting is a bad idea, of course no one listens to me because I'm not on the infrastructure team. This outage was such a nightmare, and did not win our team any points with users because we were stuck saying "I know this is critical infrastructure, but there's nothing we can do, all we can hope is your email isn't lost in the aether."

I remember a time when tech companies advertised their insane uptimes, when an outage half this size would have cost customers, but Microsoft these days doesn't have to care about their customers, and can get away with taking down entire enterprises because they haven't heard of a test environment. I would truly love to see a migration away from Microsoft and Azure, but it seems everyone is stuck because no one has the capacity to self-host these services, and even if they did, these microsoft outages would still probably take them down since cloud integration is such a monumental "feature" that companies have been pushing for years. I wouldn't have switched away from Linux on my personal infrastructure before, but I sure as hell wouldn't migrate to Micro$oft after last night.

You could try forwarding user complaints on to whoever turned out an external provider. Or just refer the user's boss to the turner-down's boss.