Your website worked yesterday. Today a visitor sends you a screenshot that says "503 Service Unavailable", or your hosting email says your account has reached a resource limit, or the page loads on a good day and fails on a bad one. This guide is for the business owner or office manager who has to decide what to do next without being technical: what the message is really saying, how to read the Resource Usage page if your plan has one, the short list of causes that explain most cases, and what to check before paying for a bigger plan or opening a ticket. It is a diagnosis guide, not a comparison of shared hosting and VPS plans or a list of signs that you have outgrown a plan. Those exist already and are linked where they apply. It deliberately gives no CPU, memory or process figures, because those differ from host to host and plan to plan, and any number here would be a guess about your account.
What the error is really telling you
A 503 is the web's way of saying "service unavailable right now". The server is reachable, but it cannot produce the page at this moment. On shared hosting, many accounts share one machine, so hosts commonly give each account an allowance for processing time, memory and simultaneous requests. When the account reaches an allowance, new requests are refused or delayed until usage falls. Some hosts show a specific message such as "Resource Limit Is Reached", sometimes with a 508 code, while others show a plain 503. The words differ, the idea is the same.
The important point is that a 503 is a symptom with several possible causes, and only some of them are about resources. Treat the message as the start of a short investigation, not as a verdict that you need to spend more.
| What you see | What it usually means | First thing to check |
|---|---|---|
| "503 Service Unavailable" on every page, at random times | The account or the server is refusing new requests for a moment | Resource Usage for faults at the same times |
| "Resource Limit Is Reached" (sometimes with a 508 code) | Your account hit an allowance set by the host | Which line in Resource Usage shows faults |
| "Briefly unavailable for scheduled maintenance" | A WordPress update did not finish, so the site is stuck in maintenance mode | A leftover maintenance file in the site's main folder |
| 503 only after a plugin or theme change | The new code is too heavy or is failing | Undo that change, then test again |
| 503 or timeouts during a campaign or newsletter | Real visitors arrived faster than uncached pages could be built | Whether caching is on; the campaign time against the fault time |
| Slow pages but no error message | Often a processing or disk limit slowing the site rather than stopping it | Resource Usage plus the speed checks in our WordPress speed guide |
Rule out the non-resource causes first
Two minutes of elimination saves a lot of confusion later. Work through these before you open any graphs.
- Is it the whole site or one part? If the home page loads but the checkout, the search or the admin area fails, the problem is a specific script, not the account as a whole.
- Does another site on the same hosting account load? If one site fails and a second one is fine, the cause sits in the first site's code or content. If both fail together, look at the account or the server.
- Was an update in progress? WordPress puts the site into maintenance mode during an update and normally removes the marker when it finishes. If you started an update and the browser tab closed, the marker can stay behind. The fix is to remove that file through the cPanel File Manager, after taking a backup.
- Did anything change just before it started? A new plugin, a theme update, a changed setting, a large import or a new integration is the most common trigger, and the easiest to undo.
- Is it your connection or the site? Try a phone on mobile data, and ask someone else to open the page. A cached copy on your own computer can make a recovered site look broken, or a broken one look fine.
If none of these explains it, and the errors keep coming back, the account is probably hitting an allowance. If the site is completely unreachable rather than showing an error, the basics are different: the first-week checklist, what to check on your hosting account, lists what is worth confirming, including where your tickets and backups live.
Find the Resource Usage page, if your plan has one
In cPanel, the home screen groups its tools under headings. Look under Metrics for a tool called Resource Usage. On hosts that enable it, it shows a snapshot of what your account used against each allowance and a count of faults, which are the moments a limit was reached. Menu wording changes between cPanel versions and hosts, and some plans do not show the page at all. If you cannot find it, that is not a sign that something is wrong: it simply means you need to ask the host for the same information.
This guide cannot tell you which of Gotka's plans show the page or what the allowances are, because neither is something we state here. If you want to know, ask in a ticket from the client portal and ask for the answer in writing. For a general tour of what cPanel contains, see what cPanel hosting includes.
One practical warning: where the page exists, it often shows only a recent window, commonly about the last day. That means the evidence disappears quickly. When you see an error, open the page that same day and take a screenshot with the time visible.
How to read it without being technical
You do not need to understand the numbers in detail. You need to know which lines can stop a page from loading and which only slow it down, and then to look at when the problem happens. Our guide to shared hosting versus a VPS explains the four main measurements in more depth. This is the shorter, symptom-first reading.
| Line on the page | In plain words | What hitting it usually feels like |
|---|---|---|
| Processing time (CPU) | How hard the server worked for your account | Pages get slower; on many hosts this slows the site more than it breaks it |
| Memory | How much working space your site's tasks needed | A task is stopped, so a page or an admin action fails with an error |
| Simultaneous requests (entry processes) | How many visitors the account was serving at the same instant | New visitors are turned away with a 503 while the busy moment lasts |
| Total processes | How many tasks the account had running in all | Errors or stalled actions, often when scheduled tasks overlap |
| Disk read and write speed | How much your site and backups were reading and writing | Slowness, most often while a backup or large import runs |
| Faults | The count of times a limit was reached | The number to match against the time of your errors |
Three habits make the page useful. First, look for faults, not just high usage: a line can sit near its allowance for a while with no fault, and a single sharp spike can produce faults on its own. Second, write down the times of the errors you saw, in the time zone you see them in, and compare them with the times on the page, because servers sometimes display a different time zone from your phone. Third, check on several different days. One spike after a newsletter is normal. The same fault at the same hour on five days in a row is a pattern, and patterns have causes you can find.
The usual suspects, in the order worth checking
Most resource problems on small business sites come from a short list. Check them roughly in this order, cheapest and quickest first.
| Suspect | How to spot it | What usually helps |
|---|---|---|
| A heavy or misbehaving plugin | Errors began after an install or update; admin pages are slowest | Switch the newest plugins off one at a time and retest; remove ones you do not use |
| Overlapping scheduled tasks | Faults at regular minutes past the hour | Reduce how often tasks run; on WordPress, consider running the scheduler from a real timed job in cPanel instead of on every visit |
| Backups at busy hours | Faults match your backup schedule | Move backups to the quietest hours, such as overnight; see our website backup guide |
| Bots, crawlers and login attempts | Spikes with no matching campaign; many requests to the login page in the access logs | Block or rate-limit abusive traffic with a security plugin or by asking the host; secure the account, see cPanel account security |
| Pages rebuilt for every visit | Faults rise with the number of visitors; the site is slow even when quiet | Switch on page caching suited to your server, so repeat visitors receive a stored copy |
| Oversized images and media | Slow pages, heavy downloads, big files in the media folder | Compress images before upload; move large videos to a video service |
| A real traffic surge | Faults line up with an advert, a newsletter or a viral post | Turn caching on before the campaign; a tier with higher allowances is the answer only if surges keep happening |
| Several sites sharing one plan | Faults appear after you add another site to the account | Find which site causes them; see multiple websites on one plan |
Bot traffic deserves a special mention. Hosts and security plugins often report that a good share of requests to a small site are from automated visitors, and a flood of login attempts can use up the allowance without any customer being affected. The cPanel access log tools can show which addresses and pages are being requested most often. This is a place to look, not a guaranteed cause, so confirm it against the fault times before acting.
Fixes in the order of cost
Once you have a likely suspect, apply one change at a time and watch for several days. If you change five things together you will not know which one worked, and you may break something else.
- Take a backup first. Download a full backup before touching plugins, files or settings.
- Undo the most recent change. If errors started with a new plugin or update, roll it back and retest. This is free and often the whole answer.
- Switch on caching. For a WordPress site, a caching plugin or the server's own cache lets repeat visits use a stored copy instead of building the page again. Our WordPress speed guide covers this, and how LiteSpeed affects speed explains why the server software matters.
- Remove what you do not use. Deactivated plugins can still be a risk; delete the ones you have finished with.
- Move heavy jobs to quiet hours. Backups, imports, image processing and scheduled reports belong overnight.
- Deal with abusive traffic. Rate-limit login attempts, and ask your host what they can block at server level.
- Re-check the page after each change. Faults should fall within a few days. If they do not, the next step is a conversation with the host or an upgrade, not a sixth plugin.
A worked example: the same error, two different answers
Imagine a small online shop that gets a 503 report from customers most evenings. The owner assumes the plan is too small. Resource Usage shows faults on the simultaneous-requests line between nine and ten at night, which is also when the shop posts on social media. A look at the plugin list shows no page caching and a stock-sync plugin that runs every few minutes. Switching on caching, slowing the sync and moving the nightly backup to the early morning removes the evening faults on the same plan.
Now imagine a second business where the errors continue after the same cleanup. Faults appear every working afternoon, caching is on, the plugin list is short and the access log shows genuine customers. That is the pattern that justifies asking the host about a higher tier. Both owners saw the same message. Only the second one had a plan problem, and the evidence that proved it was collected before spending anything. (The two shops are illustrations, not customer cases.)
When an upgrade is the right answer
Upgrade when three things are true together: faults keep returning at predictable busy times, you have already applied caching and removed the obvious causes, and the traffic in the access log is real customers, not bots. At that point the site is simply doing more work than the plan was sized for. Our guide to the signs it is time to upgrade your hosting plan goes through the other limits worth checking, such as websites, mailboxes and storage.
Before you pay, ask the host to state in writing what the higher tier allows for processing, memory and simultaneous requests, and which line of the page you expect to improve. A tier is only a fix if it raises the allowance that you are actually hitting. If your hosting is on Gotka, the tiers differ in the number of websites, email accounts and storage, as set out in the Go AI tiers compared; this guide does not state resource allowances, so ask support for the specifics.
What to send support before you wait
A clear report gets a faster, more useful answer. Include these in one ticket:
- The exact web address of the page that failed, and whether other pages work.
- Dates and times of each error, with the time zone, for example Malaysia time.
- The wording or a screenshot of the error, and a screenshot of Resource Usage if you can open it.
- What changed in the day or two before: plugin, theme, update, campaign or a new site.
- What you have already tried, so nobody repeats it.
- Your question, for example "Which allowance is my account reaching, and what does the next tier allow?"
The server-side logs show what the control panel cannot, which is why support can often see the cause in minutes once they have the time and the address.
Common mistakes
- Upgrading first. A bigger plan raises the ceiling on a site that may not be touching it; the problem returns, with a bigger bill.
- Changing many things at once. You cannot tell which change helped, and you may introduce a new fault.
- Checking only after the error. If the page keeps only a recent window, the evidence is gone by the time you look. Screenshot it on the day.
- Ignoring time zones. A one-hour or eight-hour offset makes unrelated events look connected, or connected ones look unrelated.
- Blaming the host without evidence. A ticket with times, addresses and a screenshot gets results; "my site is down again" does not.
- Leaving unused plugins installed. They add risk and workload for no benefit.
- Skipping the backup. Every fix touches files or settings; take a copy first.
Where Gotka Technologies fits
Gotka's Go AI hosting plans run cPanel on LiteSpeed servers with free SSL, daily backups, free migration and 24/7 support through client-area tickets, and a 99.9% uptime guarantee. Prices start from RM17.80 a month on the 36-month Go AI Lite plan. This guide is general advice about reading resource-limit symptoms on any shared hosting account; it does not say which pages your Gotka plan shows, does not list CPU, memory or process allowances, and does not interpret how an error would count against any uptime promise. For anything specific to your account, raise a ticket from the client portal with the details above. If you would rather have someone watch the site for you, the optional Care plans start from RM250 a month and include uptime and security monitoring, and you can decline them and look after the site yourself.
Key terms used in this guide
- 503 Service Unavailable: a web response meaning the server cannot produce the page right now.
- Resource limit: the allowance a host sets for how much processing, memory or simultaneous activity one account may use.
- Resource Usage: a cPanel tool, on hosts that enable it, showing how much of each allowance an account used.
- Fault: a recorded moment when an account reached one of its limits.
- Entry process: a visitor's request being handled by the account at one moment; many at once can reach the limit.
- Caching: storing a finished copy of a page so it does not have to be built again for each visitor.
- Scheduled task (cron): work set to run automatically at a given time or interval.
- Maintenance mode: a temporary state where a site shows an "unavailable" notice, normally during an update.
What does a 503 or "Resource Limit Reached" error mean on shared hosting?
It means the server could not handle a request at that moment. On shared hosting the usual reason is that your account reached a limit on how much processing power, memory or simultaneous requests it may use. Other causes include maintenance mode, a crashed process or a traffic flood. The wording and the numbers differ between hosts, so check your own plan's page rather than assuming.
Where do I find Resource Usage in cPanel?
Look in the Metrics section of the cPanel home screen for a tool named Resource Usage. It shows how much of each allowance your account used and a count of faults, the times a limit was hit. Not every plan or host switches it on, and menu names vary by cPanel version, so if you cannot find it, ask your host in a ticket.
Do I need a bigger hosting plan to fix a 503 error?
Not necessarily. A 503 is often caused by something your site does repeatedly, such as a heavy plugin, overlapping scheduled tasks, a backup running at busy hours, bot traffic or pages rebuilt for every visit. Fix and measure those first. Upgrade only if faults keep returning at predictable busy times after the cleanup, and ask the host to state the new allowances in writing.
Why does my site show a 503 error only at certain times of day?
A pattern by time usually points to something scheduled or to real visitors. Backups, scheduled tasks and imports run at set times, while campaigns, newsletters and social posts bring visitors at the same hour each day. Compare the time of the errors with the times in Resource Usage and your backup and task schedules, remembering to compare in the same time zone.
What should I send support when my site shows a 503 error?
Send the exact page address, the date and time of each error with its time zone, the wording on screen, a screenshot of Resource Usage if you have it, what changed just before (a plugin, update or campaign) and what you have already tried. A specific report lets support check the server side logs, which you cannot see, and answer in one reply instead of several.
Does a 503 error mean my hosting has gone down?
Not always. A 503 means the page could not be served at that moment. The whole server may be affected, but often only your account reached its own limit while other sites on the server carry on normally. Check whether a different site on the same account loads, and whether the error follows a pattern, before concluding that the host is down.
Signs It Is Time to Upgrade Your Hosting Plan (and Signs It Is Not)
Signs it is time to upgrade your hosting plan: site and mailbox limits, storage that keeps filling, slow pages, plus the false alarms to rule out first.
HostingShould You Buy Hosting Before or After Your Website Is Built?
Should you buy hosting before or after your website is built? What to sort out first, what can wait and what to ask your designer so a plan isn't left idle.
HostingcPanel Hosting Explained: What the Control Panel Actually Lets a Non-Technical Owner Do
cPanel is the dashboard behind most hosting plans, including Gotka. What a non-technical business owner can actually do with it, without calling support.