Oracle NetSuite datacenter plunges offline for a day, customers warned of data loss

Despite what Oracle supremo Larry Ellison might have you believe, Big Red's clouds do in fact go down. Yesterday some Oracle engineers' Valentine's Day celebrations were perhaps put on hold after the NetSuite enterprise resource planning (ERP) suite experienced an outage at its datacenter in Boston, USA. The downtime, which …

[John H Woods]

99.97% availablility in 12 months

So, pretty much like Office 364. This isn't nearly good enough for an ERP cloud, you don't want to be losing your ERP system for a day per year, either in a single disaster or in dozens of significant outages.

[Kevin McMurtrie]

Re: 99.97% availablility in 12 months

Many cloud providers say that there is no official downtime as long as one region remains operational with spare capacity. A single region outage is only "degraded" service.

[Scene it all]

Re: 99.97% availablility in 12 months

But you need to be running real-time replication across regions. If your business depends on constant uptime you are foolish if you do not implement this. A "backup every 30 minutes" is not good enough.

[Korev]

Re: 99.97% availablility in 12 months

TBH I'd be more concerned about half an hour of data going into the ether. At least when the system's down nothing can go missing...

[gerryg]

Captain Scarlet reference

Embarrassment, "big red"...

Or was it the mysterons wot did it?

[Korev]

Re: Captain Scarlet reference

Did they use Thunderbird for email?

[Captain Scarlet]

Re: Captain Scarlet reference

No, we used voice communicators and forms earth man.

[Plest]

Remind me again why our beloved PHBs demanded we all use cloud services? Oh yes, kickback and showing the biz customers how "with it" we are by slapping "Running on cloud technology!" on all the company bumpf!!

[CrazyOldCatMan]

'creeped'

Guys,

If you can't use British English, can you at least learn how to conjugate verbs properly?

The past tense of 'To creep' is 'crept'. Not 'creeped'

[DylanH333]

UPS goes poof!

"[...] smoke was reported inside one of our two data centers in Waltham, Massachusetts coming from electrical equipment in the power room".

Why do I feel like it was a UPS with a battery gone bad, and perhaps one of the batteries leaking hydrogen sulphide (making any electrical sparks also a fire hazard, hence cutting power entirely)?

OVHCloud lost a whole datacentre because a UPS fire resulted in the whole building burning down, and we had a near miss with one at my work.

in our case, one of the on-site IT guys walked into the server room to the smell of rotting eggs and burning plastic, and a very warm UPS.

Luckily, they managed to get it out of the building before anything eventuated, but apparently one of the batteries was very hot, had swelled, and then leaked into the base of the UPS enclosure.

I'm beginning to think UPSes are a serious fire hazard.