‘Get off of my Cloud’ Verizon tells users ahead of 48-hour maintenance outage Verizon Cloud will be down for maintenance for up to 48 hours next weekend and virtual machines hosted there will be inoperable, all to enable the introduction of unnamed improvements. At the time of writing, Verizon's Cloud Client Care page does not mention the outage, but users have emitted Tweets (such as the one below) in …
"I thought the whole idea of hosted, sorry, Cloud computing, is how amazing and reliable it is"
No idea where you got that perception from.
Typical public cloud services offer only 99.9% uptime guarantees - and only limited credits if they don't make it. The normal approach is to run critical services across multiple zones or providers - or to split between public and private clouds.
Ah, but the 99.9% SLA's probably only cover *UN*planned downtime.
As customers were informed, this is a planned downtime, and therefore does not count against the SLA's...
Well, this is how we were told to inform clients that have vhosts and FTP accounts on our racked servers... - damn, sorry, our "Cloud".
AC as I am still working for them, even though I'm trying to escape!
As customers were informed, this is a planned downtime, and therefore does not count against the SLA's...
But when were they informed? As this was planned downtime, I would assume it was in the diary and hence published and communicated to customers over a year back, hence this recent message is just the most recent of a string of reminders...
Just go to show there is a difference between being available 24x7 and 24x7x365.
Now that's one hell of an impact, and there was me thinking the cloud was the answer to all our availability and scalability problems.
I guess the fine print covers Verizon and screws the clients. VMs down 1 hour before, and presumably all up an hour or so later, but keep someone on standby just in case we finish early.
You do have to wonder exactly what they are doing and why they haven't structured their cloud to minimise this kind of thing. Who's their cloud software provider?
Well maybe.
Heinz and the amazing string company? Heinz will be supplying the new improved baked beans tins?
I guess that Verison are just using their usual dyslexic reading of service, they thought cloud was the same as cr*p. Still if they think they run a monopoly or is it a single parrot operation as in mono-polly)?
A good chance for others to do a bit of comparison advertising me thinks.
It's a big slice of the window to be eating in one go, but I guess if most of their revenue comes from contracts that allow this window, so be it. Heaven forbid that in the same SLA cycle they get an un-scheduled outage or need to apply another similar schedule.
Think of all those monitoring systems at Clients' on-site premises, busy polling stuff that's not there and flooding their helpdesks. Ew.
Assuming this is the only outage of the entire year (far fetched, but let's work with it) this 48 outage along takes them down to a rounded 99% uptime (99.45%)
Surely a new low in the industry. I can see enterprise customers even now having serious discussions as to where they're going next, along with their legal bods consulting contracts to ascertain how big a breach of contract this is.
/Paris 'cause she apparently has as much clue on how to run a cloud as Verizon do.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
