Internet Solutions downtime due to hard disc failures

Bobbin

Executive Member
Joined
Oct 22, 2009
Messages
8,392
Been wondering why I'm doing so many quotes for Azure services all of a sudden :D
 

Death_By_Ebola

Well-Known Member
Joined
Aug 18, 2016
Messages
170
Didn't this happen a few years back as well, when they were still MWeb? Lost an entire VM platform of theirs and many customers' data.
 

r00igev@@r

Executive Member
Joined
Dec 14, 2009
Messages
7,322
I wonder if it was a power related issue that triggered the crash of "hard disks"????
 

r00igev@@r

Executive Member
Joined
Dec 14, 2009
Messages
7,322
Still better than hetzner, but yea I won't host with IS either.

How can a cloud provider suffer HDD failure in 2019?
My assumption would be that the service to infrastructure dependency has not been mapped?
 

Kanniebos

Active Member
Joined
Aug 19, 2013
Messages
75
Typically IS if you would like to escalate your service request please press 1

I can recall walking around in IS DC once and counted over 40 failed drives in several cabinets
 

garp

Executive Member
Joined
Aug 2, 2004
Messages
9,013
Well at least it doesn't sounds as bad as Afrihost (I think it was them) that irretrievably lost VMs a couple of years back (at least I hope it's not).

My take - either use bare metal at somewhere like Hetzner with a rock solid cloud backup solution (the cost/performance ratio is still far, far cheaper on bare metal) or use one of the mainstream providers that have extensive inter-regional redundancy - basically Azure, AWS and Google.

Smaller or local/regional VM providers give me the heebie jeebies - that is, unless you have a very portable solution that can be redeployed to new VMs elsewhere very quickly.
 

elvis_presley

Expert Member
Joined
Sep 5, 2007
Messages
3,367
I wonder if it was a power related issue that triggered the crash of "hard disks"????

The datacenter itself should have an online UPS solution, so the incoming power should be scrubbed and smooth.

My company (smallish, 25 people) had our main UPS blow up a few years back; it totally killed half our server rack, but even our small company has a disaster recover site (different location) that we were up and running within minutes (just needed some BGP routes changed to redirect traffic) with zero loss in data.

The tools (hardware, but mainly software) available to set up these infrastructures is so good these days, there's absolutely no excuse for extended downtime. Either you've got useless staff or a bean counter that doesn't understand technology enough to allow the spend on these things.
 

cda

Senior Member
Joined
Dec 16, 2005
Messages
680
Because this is not the first time their platform has failed. Their terms say they don’t give a hoot if it does fail and neither do their support.
Azure or AWS are the future

Here's the inside scoop on IS. They have almost no technical competency left because they hired a bunch of rugby loving jock salesmen to lead the company, and that resulted in the inevitable - a giant sales target circle jerk and horribly imbalanced incentive scheme which made anyone with half a brain get annoyed and leave and rewarded the unethical morons who try to sell broadband like it's a used car.
 

Mystic Twilight

Expert Member
Joined
Dec 23, 2010
Messages
3,146
Here's the inside scoop on IS. They have almost no technical competency left because they hired a bunch of rugby loving jock salesmen to lead the company, and that resulted in the inevitable - a giant sales target circle jerk and horribly imbalanced incentive scheme which made anyone with half a brain get annoyed and leave and rewarded the unethical morons who try to sell broadband like it's a used car.

Call back to the Steve Jobs interview.
 
  • Like
Reactions: cda
Top