Technilife has moved

Slowly, as I get my services out of Google's free clutches and transfer them over to services that i'm paying for ... technilife has joined that queue.  The blog is now Here.

Yet another reason to get my arse of Google's free services

It is no secret that I am concerned about all the free services we've been slurping. I've had my own domains for more than a decade but it is only relatively recently that I am now able to run some of the more complex software from my hosting service without doing a Humphry on my bank account.

Some of my blogs have already gone from Blogger and I've tried a number of free services; the only ones that have hung around are this technical blog and also my life of a stranger blog. However, these will also be taken off Blogger when I have the time to port it all over.

I might actually decide, stuff the porting, and start afresh. We'll see how the mood takes me.

The only service that I really can't replicate at the moment is YouTube.

However, one more nail landed in the coffin today - when Google decided to walk all over Apple users over the info it slurped via their safari browsers after deliberately ignoring security settings.

Come on Jolla ... where's that bloody phone!

Another technical warning

We do need to be careful about what information we put on which devices, especially when we are trusting them with serious information and installing them in our homes.

Science fiction has long portrayed some evil computer as taking over resources for nefarious uses, but there is an even more evil, unfeeling, uncaring adversary at work here ... other human beings.

These parents who appear not to be fully technically clued up, installed a product which had vulnerabilities and then presumably didn't get themselves clued up on any updates. The net result wasn't a good one and it is worth reading the whole article on this one - http://www.bbc.co.uk/news/technology-23693460

In our modern life, the more devices we install, the more updates we have to keep track of. It is a technological rod for our own backs. If someone can gain access to our modern vehicle, then they can install a remote unit that will give them control of our steering; if they managed to develop a self destructing box then they could drive us to our doom and destroy evidence tracking them, to our death.  it is scary, but it is certainly possible.

It is getting to the point where our only defence is getting to be that no one has a reason to want to harm us.

Let's face it, someone who wanted us dead badly enough could simply come at us with a kitchen knife. They'd stand a good chance of achieving their aim after a bit of a struggle, and also of being caught, but they would have succeeded.

So my advice is to keep the number of devices down, certainly keep the number of wireless devices down, hard wire where possible, don't allow outside communication and keep a track of product updates; even if you have to make a separate e-mail address so that any potential dodgy vendors who sell our e-mails on, don't pollute our regular e-mail.

I'm not saying that we should be paranoid, but that we shouldn't take unnecessary risks and that we can't blindly trust technology. We need to keep a grip, and the responsibility, ultimately, is ours.

Pirate Bay - sailing to new shores

This article on the BBC this morning is worth a read, especially worth waking up for is what Tobias Andersson has to say about a third of the way down the page.

As he points out, three dimensional printing is almost on us. Just like telephony, when it suddenly appeared in everyone's homes, and then another revolution when they appeared on everyone's hips.

3D printing will change the whole game.

At the moment many physical things have been transferred in to digital information; books, films, music ... but that also includes footwear, bicycles, chairs ... imagine not having to go to a shop for these things, or having to get them delivered to your home, but just plugging a new cartridge in to your 3D printer, downloading the file and letting rip.

No longer will the physical pair of crocs be the product you buy, but rather the file itself. How about security; do you buy the file for one use and then see it self destruct, or do you use it time and time again to make yourself endless amounts of footwear? Also, being a design, it doesn't matter if the shop haven't got your size in stock. Some designs can be dynamically changed to match your own foot profile.

Andersson is right; a real revolution is just around the corner. Of course, we won't be seeing 3d printed shirts for a few years yet, but we are already seeing 3D printed food. If I was a manufacturer in basic goods, I'd be very concerned and planning ahead However, simply joining arms with the music and film industries to clamp down on the internet won't work. They have the chance to think ahead of the game ... after all, making alternate shoe patterns is a darn sight easier than making alternative films or music.

Time to smell the roses, guys.

(to the tune of The Muppet Show theme)


It's time to power the printer.
You clean those nozzles right.
It's time to get it printing...
Up a real good bike, tonight!

Time to think telephony

Obviously, I don't need to say where I'm working or who's systems I'm using, but there is still a lot that I can bring to the table that I've learned over the last few months.

New telephone systems have given us a fresh, new power. The power to deliver a better service to the customer, make it seem completely invisible and also keep unwanted calls off our back.

Much of what I'm going to say here has been available on telephone systems for years. A few bits and pieces add some icing to the cake, however, which takes an ordinary telephone meal and gives it a Michelin star (if such a thing were possible.)

The whole thing doesn't start with technology. It starts with us. Instead of thinking as the call being ours, we think of it being the customers.  When they are calling us, do they REALLY want US, or someone who can answer their query or solve their problem?

If we're on the other end of the phone, then that's great. We pick it up and that's job done. The important thing is what happens when we're not around. Is it really acceptable to say, "Let them eat voicemail!"? I mean, we could be away for a fortnight's holiday.

Should we expect them to have pen and paper ready when they call us, so that they can write down, "01555 (garble, garble, background noise) 6 4 *cough* 1 if it is urgent" ?

This is a decision that the team and management need to make. The bull needs to be taken by the horns.

Of course, there are Pick Up Groups; where if you hear a colleague's phone ringing you enter a code to pick it up. But how many people are conciensious enough to do that? I know of people who will not only fail to pick up a colleague's phone but actively complain that they forgot to divert it to voicemail.  Screw the person on the other end of the line who actually needs to talk with someone.

Pick up groups fail in modern systems because you're not going to hear a colleague's phone ringing on the desk in Edinburgh if you're in a hotel room in Leeds. Cloud systems should have a modern way to get around this, though.

Single number forwarding is another way to handle things, but forwarding to a single number has an important point of failure ... what if the other person is away from their desk?

Group hunt numbers are a reasonable way forward, becuase it goes around a group of people in the department until it finds someone. I mean, after this kind of treatment, if there is absolutely no one in the department who can answer the phone, then it can legitimately go to voicemail. You've done your damndest and that's about the best you can do.

There are other possibilities, however. Rather than go around the group and then to voicemail, is there some contact centre service that can take a message? Not appropriate in all situations, because if the call has already hit the most appropriate department then there is little than contact centre can do, other than to be a human voicemail; and that isn't the most efficient use of precious, expensive human resources.

But if a call does go to a team voice mail, then the whole team need to be able to see and handle a call.

So you can see how it is going to take a bit of thought to go through the process of thinking of the best way to handle these things.

This is where modern systems add a bit of precious icing on the cake. Applying rules to calls so that if you get a call from Bill or Ben, then you can have the call split out to your mobile phone, for example. That way, specific people still get to come direct to your hip if you're out of the office, but anyone who you don't know will get the general treatment.

Here, then, is your starter for ten. Time to invest a little patience in determining how you're going to handle your calls; cutting your cloth to your system. Think about what you want, then go to your telephony department and negotiate.  It makes the best use of your time as an individual and still gets your customers through to someone and their issue dealt with.

I'll wish you luck, but actually you shouldn't need it; a bit of common sense should prevail.

So what, actually, IS Unified Communications?

The long story short is if you take all the means by which we communicate, e-mail, telephony, IM, video call, "presence," and all that jazz, and mix it up in the one tin, then you have a unified communication system.

Imagine I wanted to contact Bill about an urgent customer query. I could look at my control panel, see that he was logged in to the system and was currently on the phone. That would leave me with a choice of sending an e-mail (which I could do with the click of an icon) and mark it as urgent ... or ... I could really be cheeky and send an Instant Message "Knock knock! Got a customer with a LARGE order and I need your help. Call me when you're free."

Actually, if Bill is able to juggle his communication, we could have an IM conversation while he is still on the phone to someone else ... after all, he might merely be on hold to some company or other, waiting for an agent while being assured that his call IS valuable to them! The whole thing could be stitched up in a few short exchanges; far faster than bouncing e-mails and without needing to engage voice.

Let's say I'm on the phone to Ben and we're talking through a concept and he just isn't getting what I'm saying. "OK - hang on a second." I could then flip on the video camera and hold up a diagram for him to look at. "Ah! I see what you're saying!" and that would be that. Job done. A picture worth a thousand words ... and possibly a few headbutted LCD screens in the process.

I could even be after Little Weeeeddddd.... (Bill and Ben, the flower pot men ... look it up. And no, I don't give a damn that I'm showing my age.) but it shows that Weeeeddddd isn't logged in. I could just hit the phone button anyway and trust that Weeeedddd has set up their rules to take me to a group hunt number in their absence, so I could talk to an appropriate colleague automatically without having to find another number manually.

Unified communication. Stitching it all together in one place so instead of spending time bashing our heads against a brick wall, we actually get to use the most appropriate communication method to get the job done ... fast.

And the other beauty is that if it is cloud based ... physical location doesn't get a look in to the equation. Time zone might, but that's about it.

UPDATE - Lucy Kellaway had a show transcript published on the BBC Magazine site. While she found the best communications for herself, I personally think she is wrong to assume that this is the way that the world, in general, is going. If anything, I believe it is the reverse.

So what IS the deal with cloud telephony?

Those who follow me on YouTube, know that I am now tied up in a Unified Communications, Cloud telephony installation.

But what does it mean for the individual and what does it mean for the company? It can be summed up in one word; flexibility.

There are a number of different implementations of Unified Communications and also different ways of engaging in Cloud Telephony. Without revealing any company names here, I can still bring some things to your attention.

Number one is to avoid "Unified Communications For Dummys" by Wiley publications. Put this book at such an arms length that it is stored on a book shelf on Mars. The primary author is Tony Bradley; a Microsoft bod. And it tells in the book. The whole thing is such a Microsoft, Skype, Outlook push that I almost wanted to stick my fingers down my throat on some occasions, which also included this stunning example of how not to do it on page 37...

TIP - Check out how people around the globe are using Skype in their business and learn what Skype can offer your company at http....

Wiley, you've shamed yourself this day.

At the basis of this, telephone exchanges have been charging onwards for decades, but we've still been talking with them via little boxes of buttons on our desks. Mobile phones have got smarter, but our mainstay business telephony has not followed it. Or rather, it has, but these features are usually hidden behind a confusing array of code sequences.

So we start with the soft phone itself; the solution to all this number pushing malarky. This can be a piece of software on your computer which allows you to hook up a headset device and video camera like many of us internet geeks are already used to, but also an internet based interface so we can log on to our telephone control panel wherever we are, as long as we've got internet access. Tablets, even mobile phones.

While we're talking of mobile phone software, once the soft phone is engaged on the device, it can route calls via the main system very easily, making it simple to separate personal and business call costs, as long as you're not dealing with roaming charges as well. Also, the software I've been using also allows me to control conference calls as well as my rules. (Note- this usually requires reasonable "data" connectivity as well as a basic mobile voice phone signal; something like 3G.)

Ah yes; rules. The ability to filter and handle calls just like I filter my e-mail. I can get really complex in the way things are handled, and then simply flip a profile to change the whole structure of how my calls are handled and where they are diverted to.

Diverting calls. Prefered devices. I can tell the system which end point to use. I'm not just stuck with the headset. If I'm at a different office I can pass the DDI number to the system and it will deliver my calls to that end point; even going as far as setting up a call to that number if I want to use the system to dial on, so that my host doesn't end up with the telephone charges.

To make the most of charges and enable cost saving, I've got to think, however. It is a powerful system and the onus is on me to ensure that I strike that balance of delivering the goods for the company while also keeping costs down. It puts the responsibility for communication firmly on my shoulders.

But being in the cloud frees you from physical geographical location. In fact, it frees everyone. You no longer need to have teams of people in the same physical location.  You're support teams could make the most of commute distance and only have to go to the office physically nearest to them. To the system, where they are doesn't matter. You're telephony support staff haven't got to deal with multiple forwards to get something to work, and then undo and re-do something when they move to another office for a day. It is all automatic because they don't have to be physically connected to a site system in order to register with the cloud. Heck, they could even be at home, snowed in, and still field all their business calls just like they were in the office ... even if they are a member of a complex call center team!!!

Imagine that all your staff were all snowed in; unable to get to work, but still able to work effectively. That must be an employer and an employees dream, surely?

A dream, sure, and one that requires effort if it is going to work ... on behalf of management. If the managers aren't up to the job of managing staff and work loads under such flexible working conditions, then any such business project is doomed to failure. Yahoo recently had to recall all its home working staff to the offices amid claims of some workers even running their own start-ups from home in company time. (that article is worth reading right to the end, IMHO)

The modern manager has to think beyond the nine to five and work out how the employee is to be evaluated for their contribution to the company and whether they are earning their pay cheque. If management aren't up to the task, then any attempt to reap the best rewards that improvements in telecoms can bring, are surely doomed to failure.

For example, I'm here at 14:10 typing this brief article. Why didn't I do it in my lunch hour? Simple; because someone came up to me while I was eating my salad and requested help. Once my meal was finished, instead of returning to my desk to read the news sites, I went to their desk instead and did my job. While I've got no guilt about this, a manager who is wedded to the clock could easily throw a hissy fit.

In one regard, flexible telephony has lit the blue touch paper which has lain on the ground for some considerable years. However, there is still good reason to stand well back while the hiss of flame works its way towards the gunpowder; while the PBX themselves have been slowly and steadily proving themselves over the decades, their software driven counterparts are very much the new and untested kids on the block.  Read an article by any telecoms analyst who is worth their salt and they'll probably sound the advice to be very cautious and take appropriate disaster planning measures should something go wrong.

You can't just open an office door, throw a soft phone CD in the room and let them get on with it. It is a massive jump in our approach to communication and people need to be given time and appropriate training; or else you'll find your nice, shiny CD in the shredder and everyone is still talking with the dog and bone jammed between cheek and shoulder.

Smart phone - not so smart company

What the heck has Shuttleworth been smoking?  Does he really think I would part with eight hundred American greenbacks just to get a sniff at the upcoming Ubuntu phone?

My money is on the Jolla which will cost me about two thirds of the price of the Ubuntu unit, is developed by a team that has seen it all before and promises native running of apps; potentially even a link in to the play store itself!

I've voiced my concerns about Ubuntu recently; among which is their branching out in to other areas when people using the desktop are still experiencing issues with video drivers, audio device control issues and more; many of which could be regarded as core issues which need to be put to bed if Canonical is to really gain the traction of trust in the OS arena.

Let's face it; I'm already on the hater side of the marmite debate which is Unity. With an experience like that, I'm definitely not impressed by the unit price and if that is their entry point, then I expect the Unity phone to sink without trace while Jolla's dinghy rides on.

Fragnet - have they just fragged themselves in the foot?

Horrendous incompitence on behalf of Swedish company Fragnet. (or at least, according to their web page, I believe they are Swedish; Fragnet Networks AB, Ostra Vittusgatan 36, 37133 Karlskrona, Sweden VAT Number: SE556871361301   Company Number: 556871-3613.) In February I was looking for a Minecraft server and was given a number of options by their staff...

"For what you are requesting, you may want to look at our VPS/VDS servers where you can choose the Operating System and have full SSH access to the server. The VPS servers are shared resources (OpenVZ) and the VDS servers are dedicated (KVM), both of which can be used for hosting Minecraft servers."

After buying one of their Bronze virtual packages and experiencing performance issues, they finally stated straight out...

"Your performance issues are relying on how these types of virtual servers are managing the memory and other resources being allocated to them, rather than the quantity. You could use an OpenVZ-based virtual server being allocated with 32GB of RAM and unfortunately still experience performance issues, due to how this memory would be handled. This is a recognized fact in the industry, Java applications (such as Minecraft) are never recommended to be ran on these types of virtual servers."

Apparently, this is known industry wide ... except for their own staff!!!

Despite this clearly being their own mis-selling, they refuse to budge on anything like recompense for their causing us months of heartache and the inability to open our server to other players...

"We will not give you any discount as all our VPS servers are fully self-managed and the VPS OpenVZ are not ment for high resources, as suggested if you wanted better performance and dedicated resources we can offer you a VDS/KVM the cause could of also been your configuration of your server as stated."

This is complete incompetence on behalf of Fragnet. This company is an utter disaster. Time to talk with the Swedish authorities and see if anything can be done about this pathetic company.

Also echoed in the Minecraft forum - http://www.minecraftforum.net/topic/1649803-is-fragnet-a-good-hosting-company/page__st__20

UPDATE - They offered a full refund, which I accepted, and I instructed them to take the server down. However, their "full refund" was only for part of the hire, not the full period of my receiving the service, and even THAT money hasn't hit my paypal yet. They have also taken the server down, so at the moment I have neither service OR money!

I'm not the only one having problems with them, either - http://www.gsptalk.com/topic/1876-sharing-my-view-of-fragnet-since-takeover-of-xfactor/




Further update - they don't read jack shit... and then I suffered exactly the same issue myself and had to eat some humble pie myself.

Chris Murrell || Staff
Hello Michelle,
OpenVZ can be used for Java and Minecraft, but if you wanted full and dedicated resources we would always suggest a KVM as they have dedicated resources and not shared, if you are un happy i will offer you a one time full refund.

Michelle Knight || Client
Hi Chris,
I will gladly take the refund. Please then shut down the server and kill the machine. We have all necessary files from the system.
 Once this is complete, I will determine what action I will take with regards new service.

Chris Murrell || Staff
Hello Michelle,
This has been done for you, i have terminated your server and given you the £48.90 refund back to your paypal account, i do apologise again for any issues.

Michelle Knight || Client
Chris,
Thank you, but that's a half refund, surely? I took out the service in February and then paid for a second term very recently.

Chris Murrell || Staff
Hello Michelle,
We can not refund you from then, we refunded you the last payment, thats all we can offer you.

Michelle Knight || Client
I have referred the matter to Sweden's Minister for International Trade and Foreign Affairs, Ewa Björling. When I hear back from her office I will let you know.

Chris Murrell || Staff
Hello Michelle,
Ok thats fine, please note we normally offer a 5day refund, we gave you refund out of good will and against our own policy.

Michelle Knight || Client
Hi Chris,
Just to let you know that you talk about the refund in the past tense as if you've already sent it. However, I've got home and there is nothing in my notifications.

Marcos Jimenez || Staff
Hello Michelle,
As Chris said, our return policy is of 5 days and you're out of that period why we cannot refund the invoice.

Michelle Knight || Client
The return was a good will payment outside the 5 day. Can you guys not get your story straight from one post to the other?

Marcos Jimenez || Staff
 Hello Michelle,
Our policy is pretty solid when it comes to refunds and we don't normally refund payments older than 5 days unless there's a good reason to do so.
Unfortunately the transaction is older than 60 days why PayPal does not allow us to refund the payment any more.
I've flagged this ticket for the Billing department so they can arrange with you how to proceed with the refund.
Should you need further assistance, don't hesitate to contact us.

 Michelle Knight || Client
What the hell does it take for you to read a few posts down the ticket...
---
Copy of Chris's post
---
...Chris had issued a refund out of good will and against policy. Crikey. Can you read English?

Michelle Knight || Client
Hang on ...My turn to apologise. The e-mail transfer to my mobile didn't show the whole mail.
I sincerely apologies for snapping at you and accusing you of such behaviour.
I've now read the whole post and will patiently await the billing teams response.




Update - DOS let them have it with all twenty barrels...

Posted Today, 03:16 PM ...

Fragnet, on 07 February 2013 - 11:00 AM, said:...

I find it amuzing when people who never even tried our services comments and refer to someone else (Juze). Juze posted negative information about us a year back, because he did not like TCAdmin (yet he never tried us). I do know that he was corrected my a head moderator...

With more than 200 machines running, you will bump into issues occassionally - such as DDOS attacks, network/datacenter maintenance etc. We do however monitor all our operations 24/7 (technicians working 24/7) to make sure we run optimally at all times....

We run all our servers on E3 CPUs and enterprise server hardware only. We also ONLY use low-latency premium networks, mainly because we are a Game Server Provider where latency is crucial to FPS games. We own and operate all our equipment and constantly make large investments in to improving our service and support....

We are ranked as number one on both bestminecrafthosts.com and gsprating.com - which I believe confirms that we are one of the best hosting providers out there.
...

Oh. Hell no... I'm calling BS. I am one of your former customers as of today. “Number one” host? Number one in what sense? In the sense that you define yourself axiomatically as number one and then refer back to yourself as the definition for your own imagined worth?...

Your service is broken, you know that it's broken, and you deliberately sell services you know are broken, with full knowledge that they are unfit for purpose....

Let me quote YOUR technical service email response that we have just received after months of puzzling over why a minecraft server that runs perfectly without restart for weeks on my own system, is having so much trouble running on the service your “number one” hosting sold to us......

Quote...

...Your performance issues relies on how these types of virtual servers are managing the memory being allocated to them, rather than quantity. You could have a server allocated with 32GB of RAM and still experience performance issues on OpenVZ-based virtual servers. This is more than well known in the industry, java applications (such as Minecraft) are never recommended to be ran on these types of servers....

You will easily find more information on this with a simple Google search using the following keywords : "Openvz and java", "Openvz memory management model", or "Openvz vs KVM". (KVM being used by our VDS servers)..
...

This is YOUR technical support response in an email to us......

Now go open your own website and see where you advertise OpenVZ as if it were a selling point....

As your tech support role has pointed out......

The hardware you are boasting about makes little difference for two reasons...•As your technical support role pointed out... the hardware is irrelevant if the software doesn't support the memory management required. And•Fragnet is not actually hosting its VPS services. You rent space on someone else' data farm for them. So says your website....

So, your statements about your shiny high end boxes don't actually apply and could be seen as a little deceptive....

I particularly liked the part in your tech support email response where it was our fault that your service sucks, for not googling your virtualization model and how we should have known better than to try to run a minecraft server on hosting you sold to us as minecraft server hosting. Even more I liked the part about giving actual reasons why we shouldn't have considered the service you sold suitable due to known issues with the software you are using... boasting about using....

We have had our server log spammed with “Did the time change or is the server overloaded?” usually followed by a write failure or memory error and a crash and YOUR excuse has been that you were under DDoS attack....

Let me point something out to you... This is not an excuse and it is not possible for this to cause memory and HDD access errors unless your network configuration is insanely inept....

We traceroute logged your service. We found nothing unusual from two separate routes....

DDoS can cause failed DNS queries and bad connectivity, but unless you've routed DNS requests to the physical hosts to handle AND also force your internal network to handle virtual HDD read/write traffic across the same physical network, you can't possibly expect anyone to believe that the log errors we've seen were caused by a DDoS attack that didn't appear to actually be occuring... Why, it's almost as if you were blaming a failure you expected to be noticed, on a fictional problem you didn't expect to be checked...

I'm not putting words in YOUR mouth....

Your own technical support role informed us by email that the problems we were having were our fault because we should have known better than to run a minecraft server on the OpenVZ service hosting that YOU sell explicitly AS MINECRAFT HOSTING....

Let me rub that in a bit... YOUR technical support response to us....

Quote...

...java applications (such as Minecraft) are never recommended to be ran on these types of servers..
...

Except on YOUR website where you sell exactly that as recommended specifically for minecraft hosting and your website makes it sound almost as though it's some sort of selling point... rather than a cautionary red-flag....

Yes, I have SMTP source which is verifiable for your correspondences....

Yes, I plan on beating you roundly about the head and shoulders until I tire of it, because you wasted my time trying to make a server work that you sold knowing that it was broken by the very architecture of the virtualization model you sold. After getting this charming email from your tech support desk I did as the writer suggested and did my homework... Running OpenVZ for game server hosting is absolutely nothing for you to be bragging about and it makes all of your boasts about your super high end hardware and fast network utterly irrelevant. It also explains why you seem to be omitting certain details when you talk about your services... like how a customer isn't actually getting a specific memory quota, and why using less memory will make your service suck even worse than maxing out your soft limit....

I have a theory, and it is just a theory......

My theory is that you tossed something together using software that costs you nothing......

OpenVZ is free and open source. So, you have no overhead cost for the software (and as I will point out later, you don't actually physically host your VPS services so no hardware overhead either). You sell hosting using it, knowing full well that it will suck because OpenVZ doesn't hard-allocate memory quota AND it obfuscates the fact that a portion of the memory sold is used by the host kernal and not accessible to the guest systems... It makes the memory figures appear larger than the memory the guest system can actually access. It also allows nodes with high traffic to steal the memory from lower traffic nodes because the memory is not actually assigned to a guest. Guest systems compete for memory and if your system uses too little memory, OpenVZ will give the memory you are paying for to another node with more demand. Which is one of the most important reasons why competent hosts do not use OpenVZ for game server hosting. A server doesn't have to be oversold to experience wild and erratic fluctuations in memory between nodes because whatever memory is available will be given preferentially to the nodes with the highest demand at any given moment. That means that whether you have one user or lots of users, your game is going to experience the radical swings in frame rate that we have observed, even for only one user logged in....

It has absolutely nothing to do with the imaginary DDoS attacks that don't show up in our traceroute logging... you are using a virtualization model that is unfit for use as a game server host. THIS causes the memory and HDD access errors in our log files... the imaginary DDoS attacks you use as an excuse for it... can't....

Now why would a service selling KVM dedicated hosting also sell OpenVZ when KVM is actually viable? Because you aren't actually hosting the VPS services? Because you just resell it as-is hosted by a datafarm that provides cheap cloud hosting for websites? Because you can make excuses that cover for the fact that the VPS services you are re-selling are completely unsuited for game server hosting?...

Because you can oversell the VPS services and still claim that the services you do actually host are not over sold simply by being very quiet about the fact that your VPS services are not actually hosted by you?...

I'm guessing most of the VPS customers you deal with are probably children looking for a place to set up a small private server for a few friends, and when they can't make what you sold them work, you make some excuses that don't actually make sense but sound technical to a 13 year old kid, and they eventually give up and walk away figuring that they just did something wrong and aren't smart enough to be setting up a minecraft server. You pocket their parent's credit card receipts and shrug....

It's just a theory... Maybe a theory that sounds more plausible than the frankenstein pastiche of double talk that adds up to sneaky ways of leaving out the reasons your super fast servers suck so bad....

I want to take an extra moment here and dissect your posting a bit... I wasted months on you already, so you can consider this a parting gift of a little more typing than you actually deserve....

“We run all our servers on E3 CPUs and enterprise server hardware only.”...

Which your own tech support response to us points out is irrelevant given the virtualization model being unfit for proper memory management for a gaming environment. So, let me paraphrase. By omitting the software problems from your claim, you can boast about the shiny chrome side-pipes on your hotrod and leave out the bit about the gas tank being full of water. Not precisely a lie... yet neither is it a full disclosure of the state of your services... It might give someone the impression your VPS hosting was actually worth something....

And yet... you don't actually host your VPS services (according to your own website). So, not only is your claim of shiny boxes irrelevant, it isn't actually true either....

“We also ONLY use low-latency premium networks, mainly because we are a Game Server Provider where latency is crucial to FPS games.”...

Which is irrelevant and also an overt obfuscation of relevant fact....

During our traceroute logging we found nothing unusual about your network latency....

You continued to blame erratic frame rates of 0-35 with only one user online, on DDoS attacks which you also do here in this posting... as if to say, “your service is going to suck, and we are going to blame network latency, which you probably aren't going to check...” while simultaneously boasting about your fast pipeline... because you don't figure many folks are going to check up on your claim of network latency being caused by DDoS and you figure that this claim will be an acceptable explanation for logged errors which can't be caused by network latency external to your edge router. Bad memory management however, explains everything we've seen from your service....

You seem to be simultaneously making claims to greatness based upon your network, and then blaming the failures of your virtualization model on imaginary network failures which sort of defeats the purpose of boasting about how fast your network is and does nothing at all to explain how your network layer is causing internal server errors when the server shouldn't care whether the internet is broken... especially since we found no evidence to support your claim that your network latency was unusual at all....

“...you will bump into issues occassionally - such as DDOS attacks, network/datacenter maintenance etc.”...

This statement appears to say that you know full well that your services suck and you're willing to blame that on DDoS attacks in advance, even in the same breath as claiming that you have a blazing fast network....

...which can't possibly cause the memory and HDD access errors in our logs. DDoS attacks which didn't appear to be evident while you were using them as an excuse for internal server problems they couldn't cause....

“We own and operate all our equipment and constantly make large investments in to improving our service and support.”...

Actually your own website claims that the VPS services you sell are outsourced by you, not on your shiny boxes and cheap... not high end. Let me direct your attention to the relevant page on YOUR website, http://www.fragnet.net/vps-servers.php...

“Please note to keep our prices down all our VPS Servers are un-managed by Fragnet, we only support the node and the network. “...

“We at Fragnet only utilize top of the range server machinery. We guarantee you 100% of the resources that your package contains(no overselling!). “...

You can't possibly guarantee 100% of the resources the user is paying you for on am OpenVZ host because OpenVZ doesn't support your ability to set a fixed quota, AND because the user is actually huaranteed NOT to get the use of 100% of the memory he's paying you for due to a large chunk of it being reserved for use by the host kernal and inaccessible to the guest system under OpenVZ. Any promise you make about guest node memory allocation, is a fictional statement in the context of OpenVZ....

...but as the page also states, your VPS services aren't actually hosted by you on your shiny boxes....

The statement that the services you actually host are hosted on enterprise hardware does not actually appear to be relevant to your VPS services page, yet mentioning it and your blazing fast network that magically suffers HDD access and memory errors due to DDoS attacks, makes it sound shiny too where it does not appear shiny in use. It also makes it look as though you might be deliberately confusing the features of your VPS services with the VDS services you do actually physically host to perhaps make them sound shinier than they actually are....

Now... go fire the Fragnet tech support person who told us why your services actually suck so you can carry on as usual.
 
TNB | Distributed by Deluxe Templates