Hostload Exceeded in Google Search Console: What It Means and What to Fix First

You click Test live URL or Request indexing in Google Search Console and get Failed: Hostload exceeded. No crawl, no screenshot, no verdict on the page.

The message sounds like your server fell over. Sometimes it did. More often the page is fine and the limit you hit belongs to Google's inspection tool. Those two situations need opposite responses: one says stop and wait, the other says fix your server.

Below: what the message means, how to tell which situation you are in, what to check first, and what to fix.

What Hostload exceeded means

Google's URL Inspection help defines it in one line: your site seems to be at maximum capacity for Google crawling or inspection requests, and Google can't run tests until your traffic load, as Google estimates it, drops.

"Hostload" is Google's own term. Its crawl budget documentation says Google calculates a crawl capacity limit, also known as hostload: a cap on the total time your server spends holding connections open for Google, counting both parallel connections and how long each one lasts. The point is to crawl without overloading your server.

Three details from that page explain most of what you see:

Every site starts with the same conservative limit. If there is demand and the site stays healthy, Google raises it over time.
The limit moves with your server's behavior. Consistent, fast responses push it up. Slower responses, 5xx errors, or 429 rate-limit responses push it down.
The limit is per hostname and shared across Google's crawlers. blog.yoursite.com and www.yoursite.com each have their own. Heavy demand from one Google crawler, such as AdsBot, can leave less capacity for the others.

Hostload exceeded is Google holding back before the fetch. Nothing on the page was tested, so it says nothing about the page's content, robots rules, or indexability. That is what separates it from a Server error (5xx), where Google did reach your server and got a failure back, and from a crawl anomaly, where a fetch started and went wrong.

The same wording also appears in video indexing as Thumbnail could not be crawled due to hostload: Google couldn't fetch the video thumbnail because the site seemed to be at capacity for crawling requests.

Which situation are you in?

There are two very different causes behind the same message.

Situation A: a Search Console tool-side limit

The error shows up only when you use URL Inspection's live test or Request indexing. Your pages are still being crawled, and nothing else in Search Console looks wrong.

This is what the big October 2023 spike looked like. Search Engine Journal reported a surge of the error from sites on many different hosts, seen when people used the live test and indexing requests in URL Inspection. Google's John Mueller said people were spamming the Inspect URL and Submit to Indexing tool, and that normal crawling and indexing was fine. Google's help page also notes per-property daily limits on live inspections and on indexing requests, so heavy manual use runs into walls by design.

Situation B: a real crawl capacity problem

The error lines up with signs that Google is struggling to crawl the host: Crawl Stats shows availability problems or climbing response times, crawl responses include 5xx or 429, and Page indexing shows Server error (5xx) or a growing Discovered - currently not indexed bucket. Google's crawl budget page names this case directly. If your site can't be crawled because of server capacity, "for example, you're getting Hostload exceeded in the URL inspection tool", add server resources if that makes sense for your business.

Confirm which one it is before you fix anything

Four checks, in this order. They take minutes and tell you whether there is anything to fix.

Crawl Stats host status. Open Settings, then Crawl stats. Host status covers the past 90 days in three categories: robots.txt availability, DNS resolution, and host connectivity. Green, or one old blip, points to Situation A. Recent or repeated host connectivity problems point to B.
Average response time and crawl responses. On the same report, check whether average response time has been rising, and whether the crawl responses table shows a meaningful share of server errors (5xx) or 429s. Click a row to see example URLs.
The indexed version in URL Inspection. Inspect the URL and read the Google index result, not the live test. A recent last crawl date means Googlebot is still getting through on its normal schedule, even if your manual test was refused.
Your server or CDN logs. Filter for Googlebot and look for timeouts, 5xx responses, and 429s or challenges from a CDN or firewall. Google's help page notes server errors can be transient, so a live test can fail when normal crawling would not, and the reverse.

If all four look clean, you are in Situation A. If two or more show strain, treat it as Situation B.

If it is the tool: stop re-requesting

Stop clicking Request indexing and Test live URL on the same property for a while. Repeated attempts use up the daily limits, and Google's definition counts inspection requests, not just crawling, against your host's capacity.
Let discovery do the work. Google's help page says that if you want many pages indexed, submit a sitemap rather than requesting them one by one. Link new pages from pages Google already crawls.
Come back later and inspect one or two URLs that matter, not the whole list.

A request was never a guarantee of indexing anyway. If a new page still hasn't been crawled after a few weeks, work through URL is unknown to Google and Discovered - currently not indexed rather than retrying the button.

If it is your server: what to fix first

Google's crawl capacity limit rises when your host is fast and reliable, and falls when it isn't. Fix in this order:

Server errors. Stop the 5xx responses first. Google's crawl budget page says server errors push the crawl capacity limit down.
Response time. The same page lists latency and Time to First Byte among the crawl health signals. Caching, a faster origin, and fewer slow database queries per request all help. Google also recommends supporting 304 Not Modified so unchanged pages don't need a full download.
CDN and firewall rules. Check whether bot protection or rate limiting is returning 429s or challenges to Googlebot. A 429 is a rate-limit signal, and Google crawls less when it sees one.
Wasted URLs. Consolidate duplicates, return 404 or 410 for removed pages, fix soft 404s, and avoid long redirect chains. Each of these spends crawl time on URLs that don't need it.
Server resources. If the host is simply too small for the site, add capacity. Google names this as one of the two ways to get more crawl budget. The other is content quality.

Then watch Crawl Stats over the following weeks. Response time should settle, host status should go green, and total crawl requests should recover as Google raises the limit. Google's Crawl Stats help says you can't ask Google to increase your crawl rate, so this is the only route.

What not to do

Don't hammer Request indexing. It is the most common way to trigger the tool-side version.
Don't upgrade hosting on the strength of this message alone. Confirm in Crawl Stats and logs first. If those are clean, a bigger server changes nothing.
Don't return 503 or 429 for days to steer Google. Google's Crawl Stats help says doing that for more than two or three days can make Google crawl your site less in the long term.
Don't use robots.txt to shift crawl budget around. Google's crawl budget page says blocking some pages doesn't move that budget to others unless you are already at your capacity limit.
Don't start crawl budget work on a small site. Google says sites without large numbers of fast-changing pages, or whose pages get crawled the same day they are published, usually only need an up-to-date sitemap and a regular look at the Page indexing report.

Frequently asked questions

Does Hostload exceeded hurt my rankings?

Not by itself. It means one fetch or test was skipped, not that a page was judged or demoted. If it only appears when you use URL Inspection, normal crawling usually carries on. It matters when it reflects a real capacity problem, because Google lowers its crawl capacity limit for a host that slows down or returns server errors, and new or updated pages then get crawled later.

How long does Hostload exceeded last?

Google's help page says it can't run tests until your traffic load, as Google estimates it, drops. There is no published timer. If you were only hitting the inspection tool, stepping away from it and trying again later is usually enough. If the cause is a slow or failing server, it lasts until the server responds reliably again and Google raises its limit over time.

Is my web host causing Hostload exceeded?

Sometimes, but check before you blame them. During the October 2023 spike the error showed up for sites on many different hosts, and Google said normal crawling was fine. Your host is the likely cause only when Crawl Stats also shows host connectivity problems, rising response times, or 5xx and 429 responses.

Can I ask Google to raise my crawl rate?

No. Google's Crawl Stats help says you can't tell Google to increase your crawl rate. The limit rises on its own when your site responds consistently with stable or faster response times. What you can do is add server resources if capacity is the real constraint, and stop wasting crawls on duplicate, redirected, or soft 404 URLs.

Keep the two cases apart

Hostload exceeded is a capacity message, not a verdict on your page. If it only appears when you use URL Inspection, stop pressing the button and let sitemaps and links do their job. If Crawl Stats and your logs show a slow or failing host, fix server errors, rate limits, and response times first, then let Google raise its limit.

Either way, the useful question afterwards is whether new and updated pages are getting crawled and picking up impressions. ClickFlow connects to Search Console and tracks page-level performance, so pages that are stuck are easier to spot.