Linux - ServerThis forum is for the discussion of Linux Software used in a server related context.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Please could someone assist me with a SSH query that I have
I plan on allowing 150-200 SSH connections to a Linux box and I was wondering what type of hardware and internet connection speed I would require
Each SSH connection will be transferring about 12kbps and although the maximum number of connections are 150-200, realistically I am sure that the server will only recieve about 80-120
Distribution: Solaris 9 & 10, Mac OS X, Ubuntu Server
Posts: 1,197
Rep:
hmm. You can do the math. You want a capability for a max 2.4Mbps. I don't see any reason why most hardware shouldn't be able to handle that, it's the network that's at issue, and the question is whether you are connecting from your home, a colo, or what? You're probably not going to be able to do it from home, unless you have a very unusual home setup. See http://en.wikipedia.org/wiki/Broadband_Internet_access for a summary of speeds of various kinds of connections.
Distribution: Solaris 9 & 10, Mac OS X, Ubuntu Server
Posts: 1,197
Rep:
According to top on my Sun server, I'm using about 5M+7M total size each ssh connection, but the actual resident memory varies and is less than that total. That's a root process and the user process (2 processes) that are spawned for each connection. If you have a Linux system, and have some ssh connections, try top. You can enter c while viewing top and then ssh, and it will show you just the ssh connections. Here's a sample from my Sun server:
SIZE Total size of the process (text, data, and stack) given
in kilobytes.
RES Resident memory: current amount of process memory that
resides in physical memory, given in kilobytes.
Based on your report, I will budget 12MB of RAM per user, so if I need to support 150 SSH connections to my Linux box, I will need 1.8GB of RAM plus about 200MB for a light Linux OS, I am looking at about 2GB which shouldnt be a problem
Are there any other things that I should worry about if a single box is accepting 150 SSH connections? What type of processing power would I need?
Oh and I am hoping to run this solution on a fairly powerful VPS (virtual private server) in the states... I have been told that I should rather stick with dedicated machines but they are way more expensive... What are your thoughts?
Distribution: Solaris 9 & 10, Mac OS X, Ubuntu Server
Posts: 1,197
Rep:
I would think that most server class machines could handle that without too much trouble. I've never used a virtual server, but I presume they have service agreements that tell you what resources you are guaranteed and what you will be charged for based on usage.
I hadn't offered a comment before, because you wanted feedback on the virtual server question. If someone with firsthand experience with virtual servers could jump in with some examples, that might reassure you. Otherwise, since it's a virtual server, you could look into what the scaling options are for the service you are looking at. Can you easily ask for more resources (and pay more) if you find that your first attempt is inadequate?
they have service agreements that tell you what resources you are guaranteed and what you will be charged for based on usage.
That is correct
Quote:
Originally Posted by choogendyk
Can you easily ask for more resources (and pay more) if you find that your first attempt is inadequate?
The problem is that most VPS providers only sell them in tiers or in packages. For example, plan 1 could be 2.4GHZ, 512MB RAM, 60GB HDD and then the next one would effectively double all the hardware with the exception of the processor
This is all good and well, but unfortunately this means that the price jumps quite a bit as well and in most instances, the only reason why I would take a "bigger" plan would be to benefit from more RAM, nothing else. (So far, all the providers Iv looked at do not allow you to increase memory without taking on another plan)
It seems that if I need a lot of memory, the only way I am going to get it is to have a dedicated server, which means the entire box belongs to me (and not split up into multiple XEN/etc virtual machines)
But if I can grab a suitable VPS that has enough memory, is reliable and most importantly, offers a high level of PERFORMANCE for my product, then my problem is solved because VPS's are very affordable and flexible when it comes to growth
Has anyone had good/bad experiences using a VPS for their products?
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.