Failed to find any meta tag from the page markup


We’re trying to improve our twitter posts by adding Twitter Cards. Currently we’re adding the following tags:



Running this through the validator tool ( against our test or prod site always returns the following error with nothing parsed:
“Failed to find any meta tag from the page markup”

However, if I download the page (Firefox > save as > complete web page) and then upload that to Amazon S3 and serve it from there, the validator parses it as I would expect it to. There’s no difference in the tags that the tool should be seeing between the two sites.

I’ve validated from the access log requests that Twitter is getting the same content that I’m seeing from Firefox/wget/etc.

Anyone have ideas on what’s missing?




Anyone actually follow these message boards? Any other way to get assistance?


Hi @sander_gates,

Sorry for the delay. The URL you provided might have gone stale. I tested with this one:

At a first pass, I don’t see a twitter:card tag, which tells our system what kind of tag we should display. It should look like the following:

While we fallback to open graph tags, the twitter:card tag is required for us to be able to parse your site for tags at all.

Let me know if this helps, and sorry again for the delay.


Thanks Ryan.

Yes, it looks like that link went stale (actually, the content was probably deleted at some point by the user). And thanks for picking up on the missing tag in the XvFB9 link - I’ll need to look into that.

The better example though would be - If you view source on that page, you’ll see the appropriate tags:
<meta name=“twitter:card” content=“summary”/>
<meta name=“twitter:image” content=“”/>
<meta name=“twitter:title” content="Time for some Bruins.

<meta name=“twitter:site” content=”@Fancred"/>
<meta name=“twitter:app:id:iphone” content=“614787471”/>

It is admittedly missing the description tag, which we’ll be patching soon, but the validator is complaining about missing the twitter:card tag, which is clearly there.

Late last night I did track down what I think the answer is in this case however. Or if not the cause, at least a fix. We added the following meta tag to the page:
<meta http-equiv=“Content-Type” content=“text/html; charset=UTF-8”/>
and the validator started working.

This is strange for a few reasons:
- The http headers already specified the content type as text/html
- We already had a meta tag of <meta charset=“utf-8” />
- While declaring this tag is a good practice, I don’t see it as a requirement anywhere, and all browsers I’ve tested work fine without it

So… I’m not positive what the root cause is - I’m happy to help find it if the validator can spit out more useful messages (it would be great if there were an option to spit out what content the validator actually pulled/parsed), but for now I think our problem is plugged.