• Howdy! Welcome to our community of more than 130.000 members devoted to web hosting. This is a great place to get special offers from web hosts and post your own requests or ads. To start posting sign up here. Cheers! /Peo, FreeWebSpace.net
managed wordpress hosting

How many sites can a linux / cpanel server handle?

Now lets get to the interesting part. There is a host out there called x10hosting and many others who have a huge number of users on their free hosting service. For example x10 has over 400k users. That means if they have 500 sites/users per server then they have 800 servers, that is awefully costly and impossible to run, economically speaking. Thats what challenges your argument about overloading. Because they have done it. Its real. :lol:

I am speculating that A) their hosting spans more than one machine, and B) their servers are probably extremely oversold.

It IS possible to have 5000+ users on a single machine. I've seen it first hand. Is it practical? No. Was it stable? No. Was it oversold? Extremely.
 
I am speculating that A) their hosting spans more than one machine, and B) their servers are probably extremely oversold.

It IS possible to have 5000+ users on a single machine. I've seen it first hand. Is it practical? No. Was it stable? No. Was it oversold? Extremely.

I know its smells a bit fishy but some where within that mess they have figured out a business model that works and they are successful at it. Their servers are also not too slow, i have used it on several occasions.

What enlightens me are mainly two things. If the are doing it then linux must have passed that barrier of hosting 5k or 10k sites per server. Bcs thats the only affordable figure that matches with many such hosts lately. Secondly most sites on such hosts are dead. So if the server can handle many sites then the business becomes economically viable. :wave: And for those sties which does bring in some crowd, I can bring in the cash to afford it some how.

Cant wait to setup my cpanel box and give it a try! :ninja:
 
Now lets get to the interesting part. There is a host out there called x10hosting and many others who have a huge number of users on their free hosting service. For example x10 has over 400k users. That means if they have 500 sites/users per server then they have 800 servers, that is awefully costly and impossible to run, economically speaking. Thats what challenges your argument about overloading. Because they have done it. Its real. :lol:

Or, x10 hosting has a clustered server arrangement they are using to pool the required resources. Using that they could actually do the whole job with only one maybe two such clusters, but they still would show issues with any software limitations.

I'm curious myself about this, since you have a valid point that with Linux being open source surely someone out there has a fix for the UID limitation especially if some hosts actually do have thousands of users to a machine.

But in practice by the time you actually get said thousands of users in a production environment you should be able to make enough money off of them to afford a fleet of servers and staff to manage them all. Now on the bench it might be a different story, but on the bench you can safely ignore overselling effects because none of the resources are actually being used.

I think the best way to find out is to just get in there and do it, perhaps develop a script to automate the setup and teardown so you can experiment with different stress levels to see what happens. I'd love to hear the results, because even though I am never going to oversell my stuff there may come a time when I have sufficiently large hardware to actually put thousands of people onto the same operating system install- either using clustered hardware, or a really big machine that actually has the resources to pass around.

Also, I can see where having such huge numbers of clients can make the server difficult to manage. Almost all of the daemons have to reload their config file any time they are started or restarted. Apache and MySQL in particular are likely to take a long time to do this because each account will make an entry in the configuration files.

Plus if your servers use PHP CGI or some other method of running PHP as the owning user, you'll find very quickly that you'll run out of open process IDs because each account on even the slightest of traffic will attempt to launch and maintain a worker process for PHP processing.
 
All I gotta say is, I feel sorry for your future clients.

NO, you wont. I am not gonna put 30k sites on a machine. This is just for a test situation. But I would love to successfully load a linux box with 30k sites on a test situation. Bcs that has other benefits and rewards for the future.

Pls take a peek at my next post after 10min as I am gonna ask some key questions there, which might be of interest.
 
Or, x10 hosting has a clustered server arrangement they are using to pool the required resources. Using that they could actually do the whole job with only one maybe two such clusters, but they still would show issues with any software limitations.

I'm curious myself about this, since you have a valid point that with Linux being open source surely someone out there has a fix for the UID limitation especially if some hosts actually do have thousands of users to a machine.

But in practice by the time you actually get said thousands of users in a production environment you should be able to make enough money off of them to afford a fleet of servers and staff to manage them all. Now on the bench it might be a different story, but on the bench you can safely ignore overselling effects because none of the resources are actually being used.

I think the best way to find out is to just get in there and do it, perhaps develop a script to automate the setup and teardown so you can experiment with different stress levels to see what happens. I'd love to hear the results, because even though I am never going to oversell my stuff there may come a time when I have sufficiently large hardware to actually put thousands of people onto the same operating system install- either using clustered hardware, or a really big machine that actually has the resources to pass around.

Also, I can see where having such huge numbers of clients can make the server difficult to manage. Almost all of the daemons have to reload their config file any time they are started or restarted. Apache and MySQL in particular are likely to take a long time to do this because each account will make an entry in the configuration files.

Plus if your servers use PHP CGI or some other method of running PHP as the owning user, you'll find very quickly that you'll run out of open process IDs because each account on even the slightest of traffic will attempt to launch and maintain a worker process for PHP processing.

Ok there might be a few newbe queries that you or others here can answer me.

First what is this linux UID about? Does it put a barrier on how many sites a cPanel box can have? Why ? Any idea?

Even if there is any such limits the free hosts have bypassed it some how. even the big free hosts dont look like companies with a supercomputer. I woudnt say they have more that pc servers, looking at their sites! I believe that is finely doing the task for them. so no sweat.

I need to know a little more about how cpanel and linux handle sites? Please shed me some light. Does every site consumes or reserves some memory when active? That would ruin they day bcs not many sites can be active at the same time then. However if sites shares the memory and dont reserve for each site then we can handle huge traffic there.
 
:fangel: I do respect your values and that's why I love sharing my ideas and also learn your tactics. But then again, when you think from the scratch , why cant a linux box with cpanel on it create thousands of sites on it, assuming none is getting any visitors. I know that widows cant do it earlier due to file system limitation and few other things. Then why cant linux do it? The reason bugs me.

I am not saying linux cant do it. So I will soon start a test run on a linux box.

Now lets get to the interesting part. There is a host out there called x10hosting and many others who have a huge number of users on their free hosting service. For example x10 has over 400k users. That means if they have 500 sites/users per server then they have 800 servers, that is awefully costly and impossible to run, economically speaking. Thats what challenges your argument about overloading. Because they have done it. Its real.

No lol's allowed. I once disagreed with minor points of X10 Policy but they are still here.

Let's suppose they can buy hardware at $500 on bulk /special discount. Then sure, they might have 400K users.

I kept saying that the Free Host industry blinds itself with Small Dollars, so that any random company with $10,000 to burn can rewrite this industry.
 
No lol's allowed. I once disagreed with minor points of X10 Policy but they are still here.

Let's suppose they can buy hardware at $500 on bulk /special discount. Then sure, they might have 400K users.

I kept saying that the Free Host industry blinds itself with Small Dollars, so that any random company with $10,000 to burn can rewrite this industry.

You sound like its easy to get clients for free hosting? I thought it was a tough market at the moment!
 
You sound like its easy to get clients for free hosting? I thought it was a tough market at the moment!

Most people with real sites use paid hosting these days. They don't mind spending $2-$5 p/m on a website if it's actually popular and active.

The most you're going to get in free hosting for your first year is maybe one or two real sites and around 100-200 idle sites that do nothing other then gather dust. That's not including the, oh, 3000+ spam accounts you're going to have to sort through in the first two months alone. I know this from first hand experience.
 
Most people with real sites use paid hosting these days. They don't mind spending $2-$5 p/m on a website if it's actually popular and active.

The most you're going to get in free hosting for your first year is maybe one or two real sites and around 100-200 idle sites that do nothing other then gather dust. That's not including the, oh, 3000+ spam accounts you're going to have to sort through in the first two months alone. I know this from first hand experience.

Yeah, I am pretty much aware of the nature of the free hosting business. Spam spam & more spam!

Thats why I was investigation the potential of the linux boxes limit to understand if it would handle volume to survive under the dust.
 
You sound like its easy to get clients for free hosting? I thought it was a tough market at the moment!

It is.

Tao's point is that sometimes it seems like hosts starting out with pocket change end up forever tied to small change practices that in the long run cost them more because they miss out on the economy of scale.

Someone who had a bit of a reputation to go with and suddenly came up with $10,000 to invest in monster hardware upgrades probably could make a substantial jump in profitability without sacrificing reliability and performance. But most hosts these days build up slowly from small beginnings, and through that steady effort can't afford the lump sum investments that might in fact make them more cost effective.

I think the biggest thing you're seeing is people are misunderstanding you to believe that you're going to do this with live clients. I somehow doubt that is the case, you're doing this on a test bench as a case study just to see what happens. There are no actual clients that will be on this 30,000 account server, just dummy accounts for testing purposes to see what happens. If you don't run into daemon stability or UID/PID shortages, you might actually make your goal with this.

But doing what you're testing on a live production server with actual clients is not recommended by any means, as it is far too unreliable to be worthwhile.
 
This would all depend on the specs of your server and bandwidth. It could hold 50 or it could hold thousands. It also depends on how big the websites you host are and how much space a customer uses.
 
It also depends on how big the websites you host are and how much space a customer uses.

No, not really. You can host huge accounts that are 100GB each if you wanted to. The thing is, if there's no activity on any of them and they're dormant, how would that affect performance? The answer is: it wouldn't, only UNLESS you're nearing the physical capacity of the drive(s).
 
Back
Top