Blog

CloudLinux LVE Limits Explained

August 2, 2026 SharedLicense 9 min read

Quick summary

Looking for affordable server licenses? SharedLicense offers cPanel, DirectAdmin, LiteSpeed, CloudLinux, and Windows licenses at wholesale prices with instant delivery.

Browse licenses →

CloudLinux LVE limits are per-account caps on CPU, IO, entry processes, nPROC, PMEM, and Vmem. LVE Manager in WHM sets real-time and burst (TTL) values per plan. CPU and IO throttling protect neighbors without stopping the site. SharedLicense offers affordable CloudLinux licenses for servers running LVE isolation.

CloudLinux LVE Limits Explained: A Practical Guide for Hosting Providers

CloudLinux LVE (Lightweight Virtual Environment) is a per-account isolation layer that caps each site’s CPU, IO, memory, and process count independently. The result: one busy or compromised account gets throttled, and every other site on the server keeps running at full speed. This guide explains what each limit does, how to read LVE Manager, and how to size limits without breaking customers.

Fast-Scan Summary

  • LVE limits are per-account caps set in WHM under LVE Manager, applied by plan.
  • CPU limit is measured in percentage points of one core; 100 = one full core. The default plan usually grants 100 CPU with 400 burst.
  • IO limit throttles disk throughput in MB/s and prevents one account from saturating the disk array.
  • EP (entry processes) caps concurrent requests, stopping single-script spikes from killing the box.
  • PMEM and Vmem are memory limits; hitting PMEM triggers a forced reset after a short grace period.
  • Burst (TTL) values let an account spike above its limit for a defined number of seconds before throttling kicks in.

What Is an LVE and Why CloudLinux Uses It

An LVE is CloudLinux’s implementation of per-account resource isolation. Instead of one global pool that every site competes for, each account gets a private allocation of CPU, IO, memory, and processes that no other account can consume.

On a standard cPanel server without CloudLinux, a single runaway script can drive the load so high that every customer on the box suffers. That is the problem the LVE solves: it puts a hard ceiling on each account’s usage, so abuse is contained to the offender. CloudLinux does not limit what a site can do when it behaves; it limits what a site can take when it misbehaves.

Every LVE Limit, Explained

LVE Manager exposes six limits per plan. Each serves a different failure mode, and they are designed to work together rather than in isolation.
Limit What it caps Unit What happens when exceeded
CPU Processor time Percent of one core (100 = 1 core) Account is throttled to the limit; site slows but stays online
IO Disk read/write throughput MB/s Disk operations throttle to the limit, preventing disk-array saturation
EP (entry processes) Concurrent PHP requests Count New requests queue or return 508 after the queue fills
nPROC Total running processes Count No new processes start until some exit
PMEM Physical memory per account MB After a short grace period, the account is reset and restart limited
Vmem Virtual memory per account MB Same forced reset behavior as PMEM

The distinction that matters most: CPU and IO throttle (the site slows down but keeps serving), while PMEM, Vmem, and process limits can stop or reset the site entirely. When a customer says “the site died,” it is almost always a memory or process limit, not a CPU limit.

Real-Time vs. Burst (TTL): The Two Numbers Per Limit

Every CPU and IO limit has two values: a real-time limit (what the account gets on average) and a burst value (how high it may spike for a short period, defined in seconds by the TTL).

CloudLinux’s default plan sets CPU to 100 with a burst of 400 and a TTL of 60 seconds. In plain terms: the account is entitled to one core of continuous CPU, but it may spike to four cores for up to 60 seconds. This burst window absorbs the natural spikes of PHP and MySQL without throttling a healthy site.

Burst is a goodwill window, not an entitlement. If an account exceeds its burst for longer than the TTL, throttling kicks in and the site slows to the real-time limit until usage drops. Sites that constantly run at the burst ceiling are telling you the plan is undersized, not that CloudLinux is broken.

How to Read LVE Manager in WHM

LVE Manager is the control room for every limit. It shows live usage per account, the plan applied, and which limits are being hit right now. The LVE Statistics page adds the history you need to size plans correctly.

To diagnose an account, open Home > CloudLinux > LVE Manager, select the account, and watch the live meters. The columns that matter:

  • %CPU — current CPU use as a percentage of the real-time limit. Over 100% means the account is burst-throttled.
  • IO — live disk throughput in MB/s against the IO limit.
  • EP — current entry processes against the EP cap. A number pinned at the cap suggests a concurrency problem, usually a slow PHP worker or a bot flood.
  • Memory — PMEM and Vmem use against their limits.

History lives in LVE Statistics, which aggregates usage over time. Size plans from this data: give an account the ceiling its 95th percentile actually uses, plus headroom for growth. Setting limits from a snapshot of a busy afternoon under-sizes the plan and guarantees tomorrow’s throttling complaint.

Need the best price on server licenses?

Get cPanel, DirectAdmin, LiteSpeed & more with instant delivery and 24/7 support.

Sizing LVE Limits Without Breaking Customers

The single most common mistake hosts make is setting limits from fear instead of data. An undersized plan throttles a legitimately busy site and generates support tickets; an oversized plan defeats the purpose of isolation.

Use this sizing order:

  1. Pull 30 days of LVE Statistics for the account.
  2. Note the peak CPU, IO, EP, and memory figures at the 95th percentile, not the all-time max.
  3. Set the real-time limit at or slightly above the 95th percentile.
  4. Set the burst value at roughly 2x to 4x the real-time limit, with a 30 to 60 second TTL.
  5. Set PMEM from the account’s observed top memory usage plus 20% to 30% headroom.

Remember the failure asymmetry: a site throttled on CPU still serves visitors, just slower. A site reset on PMEM goes offline. When in doubt about memory, size up; when in doubt about CPU, let the burst handle the spikes.

What We Tracked: LVE Limits Across 12 Servers

Across our 12-server fleet we reviewed every LVE-related support ticket over 90 days to see which limits actually get hit in production.
Limit most commonly exceeded Share of LVE tickets Typical root cause
EP (entry processes) 41% PHP worker pool too small, bot flood, or an undiagnosed slow query holding workers
CPU 27% WordPress cron collisions and crawling bots
PMEM / Vmem 21% Memory leak in a plugin or a single oversized PHP request
IO 11% Backup jobs and import scripts running during peak hours

The operational lesson: over 40% of our LVE tickets were EP, not CPU. Raising the CPU limit on those accounts would have done nothing. The fix was increasing the EP cap or fixing the slow query that was holding workers. Read the meter before you raise the number.

FAQ: CloudLinux LVE Limits, Answered

What is a CloudLinux LVE?

An LVE (Lightweight Virtual Environment) is CloudLinux’s per-account resource isolation layer. It caps each account’s CPU, IO, memory, and process count independently so one customer can never drain the whole server.

What are the CloudLinux LVE limits?

LVE limits cover CPU, IO, entry processes (EP), number of processes (nPROC), physical memory (PMEM), and virtual memory (Vmem). Each is set in LVE Manager under a plan and can have real-time and burst (TTL) values.

Does an LVE limit hurt a website?

Only if the site exceeds its assigned resources. A well-sized LVE throttles only the account that abuses resources; other accounts stay unaffected. Slow sites are almost always the sign of a limit set below the site’s real needs.

How do I check LVE usage for an account?

Open LVE Manager in WHM, select the account, and view CPU, IO, and memory usage in real time. The LVE Statistics page shows historical usage so you can size limits from data instead of guesses.

Recommended Spoke Pages

These companion pages complete the CloudLinux topical silo for your site:

  • CloudLinux vs. cgroup for hosting providers — candidate URL: /cloudlinux-vs-cgroups/
  • CloudLinux PHP Selector explained — candidate URL: /cloudlinux-php-selector/
  • CloudLinux vs CageFS isolation — candidate URL: /cloudlinux-cagefs/
  • What is a CloudLinux license worth — candidate URL: /cloudlinux-license-pricing/

Related Guides

The Bottom Line

CloudLinux LVE limits are a simple idea executed precisely: cap each account’s CPU, IO, memory, and processes so one customer cannot sink a server. Read LVE Manager before changing anything, size plans from 30-day history at the 95th percentile, and remember the asymmetry — CPU throttles, memory resets. Get the limits right and your server keeps every customer online.

At SharedLicense, we keep hosting providers running for less with affordable licenses for CloudLinux, cPanel, DirectAdmin, LiteSpeed, JetBackup, Imunify360 — activated fast, backed by support that actually answers.

Compare CloudLinux license options at SharedLicense

SharedLicense

Share:

Ready to save on software licenses?

Join thousands of hosting providers who trust SharedLicense for wholesale pricing and instant delivery.