-
Notifications
You must be signed in to change notification settings - Fork 9.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Largest Contentful Paint element — Error! in PageSpeed #15075
Comments
I was able to repro. Looks interesting, this is from how we calculate TTFB in the LCP phases table:
|
Yes indeed i have the same issue here too. |
In PSI the network response protocol event has {
"method": "Network.responseReceived",
"params": {
"requestId": "9F23D3E13238FFE8AB51366468E20311",
"loaderId": "9F23D3E13238FFE8AB51366468E20311",
"timestamp": 381416.814998,
"type": "Document",
"response": {
"url": "https://example.com/",
"status": 200,
"statusText": "OK",
"headers": {
"X-Original-Content-Encoding": "gzip",
"X-Original-Content-Length": "648",
"X-ProtocolIsH2": "true",
"X-RequestMs": "10",
"X-ResponseMs": "1",
"X-TCPMs": "21",
"X-TotalFetchedSize": "1022",
"X-TotalMs": "31",
"accept-ranges": "bytes",
"age": "237871",
"cache-control": "max-age=604800",
"content-length": "1256",
"content-type": "text/html; charset=UTF-8",
"date": "Mon, 15 May 2023 17:59:06 GMT",
"etag": "\"3147526947+gzip\"",
"expires": "Mon, 22 May 2023 17:59:06 GMT",
"last-modified": "Thu, 17 Oct 2019 07:18:26 GMT",
"server": "ECS (dna/63A8)",
"vary": "Accept-Encoding",
"x-cache": "HIT"
},
"mimeType": "text/html",
"connectionReused": false,
"connectionId": 0,
"remoteIPAddress": "",
"remotePort": 0,
"fromDiskCache": false,
"fromServiceWorker": false,
"fromPrefetchCache": false,
"encodedDataLength": 491,
"timing": {
"requestTime": 381416.777226,
"proxyStart": -1,
"proxyEnd": -1,
"dnsStart": -1,
"dnsEnd": -1,
"connectStart": -1,
"connectEnd": -1,
"sslStart": -1,
"sslEnd": -1,
"workerStart": -1,
"workerReady": -1,
"workerFetchStart": -1,
"workerRespondWithSettled": -1,
"sendStart": -1,
"sendEnd": -1,
"pushStart": 0,
"pushEnd": 0,
"receiveHeadersEnd": 36.091
},
"responseTime": 1684173546352.936,
"protocol": "http/1.1",
"alternateProtocolUsage": "alternativeJobWonWithoutRace",
"securityState": "secure"
},
"hasExtraInfo": false,
"frameId": "9658666C7406862CD71FDB16EB0818C1"
},
"targetType": "page",
"sessionId": "2868D5FDEC09D185A870591FEEA4EE6E"
}, |
I see the same regarding PageSpeed Insights vs local Chrome Lighthouse. Error using PageSpeed but Lighthouse in Chrome shows LCP just fine. |
i don't think this should affect the SEO ranking as it should be based on real user data (finger crossed haha). Meanwhile it's true that lighthouse in chrome seems to work correctly and since this tool is just a wrapper of it, sound like the error should be within the code at least. I guess that for now (until a fix appear) if we want to know our current page-speed we can use the lighthouse devtool, am i correct @adamraine ? |
For me, it appears I fail Core Web Vitals using Search Console because of this issue. |
The audit experiencing this error is only a diagnostic used to report the LCP element, it does not directly affect your LCP value, Lighthouse score, or real user data.
This is correct. The only Lighthouse client we are seeing this in is PageSpeed insights. You can use Lighthouse in DevTools or the CLI until the issue is fixed.
Lighthouse results are not used to determine your CWV ranking in search console. |
I can confirm this is happening for us as well (as I'm sure it is for everyone using PSI). In the JSON returned from the test we do see the LCP timing and score:
However the LCP element data shows an error:
|
@WalrusSoup that is indeed a separate issue, please file a new bug. |
This comment was marked as outdated.
This comment was marked as outdated.
Appreciate the communication @adamraine ! |
The root cause here is that network requests |
A fix for this has been rolled out / is currently rolling out now (we disabled the offending code in the audit, but will re-introduce the phases table with proper support soon). |
We are seeing this now in Page Speed Insights. It popped up after we fixed a report that the content responsible for LCP was loaded lazily. We removed the lazy loading and now are consistently getting this error. |
@synth can you share what URL you are testing on? |
We made some separate fixes and now that item is no longer showing up for us, so we're not able to reproduce it anymore. Will post back if it shows up and it can repro'd for you. Thanks. |
Closing as fixed, if anyone sees this again please let us know. |
I see it quite often, but inconsistently |
Actually, same. We've seen it pop up again since the previous time, but inconsistently. |
I can repro this, will investigate |
FAQ
URL
https://web.dev/
What happened?
"Largest Contentful Paint element — Error!" is appearing on all websites, including your own site (https://web.dev/). It appears to be a bug with the latest version of Lighthouse (Lighthouse 10.2.0) when using PageSpeed Insights.
What did you expect?
The error not to appear.
What have you tried?
Tried multiple websites with no plugins, no Cloudflare, etc. It happens regardless of the setup.
How were you running Lighthouse?
PageSpeed Insights
Lighthouse Version
Lighthouse 10.2.0
Chrome Version
No response
Node Version
No response
OS
No response
Relevant log output
No response
The text was updated successfully, but these errors were encountered: