What is Unlimited Hosting? Is Everything Really Unlimited?

HomepageArticlesHosting & ServerWhat is Unlimited Hosting? Is Every...

Hello; With the concept of Unlimited Hosting, we wanted to share information in detail in order to eliminate the misunderstandings we have experienced or experienced so far, referring to words such as "Unlimited" - "Unlimited" that cause misunderstandings and misunderstandings.

What is Unlimited Hosting?

Unlimited hosting is a shared hosting package created for our customers to receive service without limiting disk and traffic.

With the unlimited package, you can create hundreds or even thousands of pages of any size.

You do not have any restrictions on disk or traffic.

However, there is only one thing that has a limit about the disk, that is the number of files you have hosted in your FTP space. Generally, it is the INODE limit in hosting companies and this limit is limited to 90% (100.000) files.

The part you need to pay attention to here is that there is no limitation in terms of disk space. For example, you can host 100,000 files in 10MB files each, 100,000 files of 1KB each.

What are the other limitations?

What is CPU Limiting?

The part written as SPEED is the part where CPU - Processor is limited and it is calculated as a single core head.

This allows hosting companies to set roughly the same performance level limits on different hardware using a single setting.

What is Virtual Memory Limit?

The virtual memory limit corresponds to the amount of memory operations that can be allocated in LVE. You can view individual process virtual memory usage by watching the VIRT column for process.
When the process tries to allocate more memory, CloudLinux checks if the new total virtual memory used by all processes in LVE is more than a limit set.
In such a case, CloudLinux prevents the allocation of memory and increments the fVMEM counter. Usually it causes the operation to fail. For CGI / PHP scripts it usually causes 500 and 503 errors.
Note: It is recommended that you disable VMEM limits (set to 0) on your system, as they are deprecated on CloudLinux 6 and 7 system and can cause unexpected problems.

What is Physical Memory Limit?

Physical memory limit refers to the amount of memory used by customer's processes.
Because similar processes (like PHP) share most of their memory, physical memory usage is generally much lower.
In addition, physical memory includes customer-used shared memory and disk cache.
In the case of disk cache - if the user does not have physical memory, the memory used for the disk cache will be freed without causing any memory errors.
When the LVE exceeds the physical memory limit, CloudLinux first frees the memory used for the disk cache, and if that is not enough, it terminates some processes within this LVE and increments the fPMEM counter.
This usually causes the web server to present 500 and 503 errors. Physical memory limit is a better way to limit memory for shared hosting.

What are IO Limits?

IO limits restrict data entry and data output in shared hosting.
It is in KB / s. When the limit is reached, processes are throttled (put to sleep).
This makes sure that processes within LVE do not cross the limit.
They work slower when the limit is reached.
IO limits will only affect DISK IO and will have no effect on the network.
It also does not consider any disk cache access.

What are IOPS Limits?

IOPS limits limit the total number of read / write operations per second. When the limit is reached, read / write operations stop until the current second expires.

What is Entry Processes Limit?

Entry processes limit the number of entries LVE.
A process always enters LVE and the counter increments by 1.
Every time the process ends, it exits LVE and the counter is decreased by 1.
We do not count the processes created in LVE itself. Also known as the ‘Apache concurrent connections‘ limit.
When there is a new HTTP request for CGI / PHP, the transaction input enters LVE.
This limit was created to prevent DoS attacks against the web server. One of the most popular attacks is to hit some slow pages on a server, dropping all Apache, Litespeed connections or putting the server off.
Once all Apache slots are used, no one else will be able to connect to the web server and this will not appear.
To solve this, we limited the login processes (commonly called concurrent connections).
By limiting the number of concurrent connections with Apache, it will cause the web server to serve 508 pages (Resource Limit Reached), which will exceed the number of concurrent requests for the site.

What is Number of Processes Limit?

NPROC checks the total number of processes and threads in LVE. Once the limit is reached, a new process cannot be created (until another one is terminated). When this limit is reached, the NPROC counter is incremented. In such a situation, your website may show a 500 or 503 error and your site will not open.