@dafyre said:

@scottalanmiller said:

@dafyre said:

When the panel is working, I DO still have access to power on / off.

Does it do anything? Or is it a marketing button?

I should have thought to check it... but I was actually just waiting around for my system to come back up so I could make backups.

A couple of us checked it. It does nothing.

@thecreativeone91 said:

It's funny noticing everyone on Twitter posting congratulating them on how well they handled this outage. Really? that's considered handling it well - especially when you had the resources to fix the issue the whole time with redundant links.

Yeah, the degree to which they failed is unforgivable. This was an intentional outage, not a mistake. They chose to keep it down all that time. And now, the servers that are back up are broken. No power cycles, no re-images. They have effectively lost the service completely.

Time for a reload..

It give me practice anyway...

They sent this update

To further clarify as I believe the meaning may have been lost in my phrasing (I apologize), the error you're receiving on boot indicates that SELinux was for some reason disable improperly (ie. not with the config files -- If the kernel looks for SELinux but it is gone, it will cause kernel or init halts, or even full-blown kernel panics.

Not sure how it would get disabled or changed without using the config files?

This is pretty bad.

60% Off BD3 today:

Cloud At Cost System Message:

04/15/2015 10:22 AM

CloudAtCost is offering 60% off today on BigDog 3’s

QTY is limited.

Package: BigDog 3

Discount: 60%

Coupon: ydA4e8USUq

@thecreativeone91 said:

Dang. I still think These IO issues caused some corruption on my CentOS box somehow.

Very possible.

One of my downloads was over 4Mb/s.

@scottalanmiller said:

At least there was a solution.

Albeit not a great one.

Congrats

Gotta say though I'm still very interested in the company,

I was really hesitant about a year ago and refused to jump on board, now that I'm on I'm happy I'm on.

I'm only excited to see where the company is going and hope they dont pull the lifetime servers (hopefully ever) for a very long time

Not sure how they handle the 1:1 Nat mapping of those IPs at the OS level.. Or I guess they might just add a new Nic for each IP.

@thecreativeone91 said:

@thanksajdotcom said:

@scottalanmiller said:

@thanksajdotcom said:

Ok, I'd think it'd be a simple thing to add the CentOS 6 repos, run a yum update, and allow the packages to update. What makes it not that simple?

That it does nothing. A CentOS 5 system, pointed to CentOS 6 repos will simply see the as not applying. You can't change your OS version using YUM.

Oh, I was not aware of that.

Maybe you are thinking of Minor release updates, not major release.

Yeah, must be.

@Aaron-Studer

Gone this all done as well thanks for the tip.

@Aaron-Studer said:

You can thank me later 😉

Hi later.

@scottalanmiller said:

My guess is that some combination of MySQL and Apache were unable to cache all of their content. You had memory being maxed out, but not over taxed, so those subsystems were not "expanding into free memory" to leverage RAM. So they were running from disk instead of from memory.

The additional of an extra CPU helps too. Rarely do workloads run ideally from a single thread. Having two threads lets there be a lot less context switching (a LOT less) while getting better cache hits.

Ok, that makes a lot of sense.

Our adult entertainment division is legally based out of the Netherlands and we still charge USD for everything, because even so most of our customers are from the US or are familiar with the currency.

CloudAtCost has completely revoked their "free for life" products that people have already paid for in full and is now simply charging for them - even ones that customers already own. Customers are now being extorted to pay annually for something they were sold specifically so that they would not have to pay for in this way. The service is no longer a scam on the technical level, it is now a full scam on the billing and marketing level as well.

https://mangolassi.it/topic/14009/cloudatcost-turning-to-extortion

It should be "All your minions are belong to us"

all-your-base.jpg

@scottalanmiller said:

@BMarie said:

I want mine for databases, software enhancement & design for websites....etc And what ever else I tickles my fancy. 🙂

The NTG lab has a Redis and MongoDB cluster. And not multiple instances on a single VM, which you can do in a pinch, but three dedicated VMs each running one instance of the DB so it is a real cluster with failover that can be tested.

I'd like to play around with it sometime.