Entry processes measure how many things are entering your account at the same time: each request to a PHP page, each scheduled task, each SSH session. They do not measure visits per day. That is why a site with few visitors can hit the limit if each request takes a long time, because slow requests pile up on top of one another. When the limit runs out, the next requests are refused and the visitor sees error 508.
When it runs out, and what you see
| What you see |
What is happening |
| Error 508 (resource limit) for a moment, then the site is back |
Requests stopped fitting and the extras were refused. When the slow ones finish, everything returns to normal. |
| It happens at the same hours |
A scheduled task, a backup or a campaign. Look for what runs at that hour. |
| You cannot reproduce it |
Normal: the limit is reached in bursts, and the owner almost never sees the error. See website errors explained. |
What fills the entry processes
| 1 |
Slow requests. A badly written query, a plugin calling an outside service, an import. An eight-second request holds its slot eight times longer than a one-second one. See what eats CPU on a website.
|
|
| 4 |
Overlapping scheduled tasks. One that takes longer than the interval between runs starts piling up on itself.
|
|
| 5 |
Many people at once, which is the most obvious case and the least frequent on small sites.
|
|
Who arrives first: the access log
To find out who fills the account, open the raw access log in Metrics, which shows each request with time and address. If the same address, or the same file, appears hundreds of times in the hour of the problem, that is the suspect. A file called too many times is usually a plugin; a repeated address is usually a robot.
|
Moving up a plan does not always fix it. More room delays the problem but does not correct a slow request: the cause returns, later and dearer. Deal with the cause first; the rest is in what to measure before you change plan.
|
|
What usually lowers the number of processes fastest: switching page caching on, so most requests do not run PHP, and moving heavy tasks to quiet hours. See LiteSpeed Cache for WordPress: the first settings. In Resource Usage a graph with this name, or a similar one, may appear, depending on the server’s configuration.
|
|
Seeing error 508 and not sure where it comes from? Tell us the time it happened and we will look at the account with you.
Open a support ticket
|
Where to see it in your account
Open Resource Usage in the Metrics section of cPanel. The graph of entry processes (it carries that name, or a similar one, depending on the server) shows how many were in use against the limit of your account, hour by hour. When the limit is reached, the next requests are refused and the visitor sees error 508; as soon as the slow requests finish, the site returns to normal. The moments when that happened are marked on the graph, and you can match them with the hour of an import, a backup or a campaign. The figure that applies to your plan is the one on that screen. If it is reached on ordinary days, even after you deal with the slow requests, you can move up a plan without losing the site.
RECOMMENDED PRODUCT Web hosting with cPanel Domain and SSL included, daily backups and the panel you already know. from R118.80/mo (3-year plan, with coupon) See plans |
Angola
Brasil
Kenya
Moçambique
Nigeria
Portugal
South Africa
United Kingdom
United States
International
