29

I get the error "Couldn't load 23 page resources" when doing the mobile friendly test at

https://search.google.com/test/mobile-friendly

However, the error message is (the rather unfriendly) "Other error."

When I visit the site in Chrome using Developer Tools and setting the device toolbar to "Responsive," it works fine with no errors, and when I go the Google Search Console and do Crawl->Fetch As Google, I get no errors.

Improve this question
10
  • 1
    If you want help with the issue, I believe it would be easier on Google's Webmaster Central Help forum, because this will likely require lots of analyzing your site and asking you questions about lots of stuff before reaching any conclusion. // If it gets resolved, you could then edit this question if needed and answer it yourself to make it useful for other people.
    – Oskar Skog
    Jul 19, 2017 at 19:51
  • 1
    ... Or a different webmasters' forum. But I can mention two reasons why WMC is the right choice: access to Google through Top Contributors and Rising Stars, and I know there's "no" SEO scum there. (They exist, but get stopped before they cause any harm.)
    – Oskar Skog
    Jul 19, 2017 at 19:54
  • 1
    Scott, I use both tools and there are indeed discrepancies in the number of issues and how they fetch. I prefer GSC's native UI, because your site is verified and it knows your site a little better than MFT. That said, MFT is still worth examining. Check for issues like blocked resources and heavy images and font files. I'd take care of any GSC issues first. For a list of possible (non-robots.txt related) "other errors," check out their documentation: support.google.com/webmasters/answer/6352293?hl=en . Good luck!
    – Henry Visotski
    Jul 19, 2017 at 21:15
  • 2
    @HenryVisotski: It looks like it was intermittent. The "other error" applies to timeouts (tested) so it's reasonable that it would apply when Googlebot decides to not even try as well ("temporarily unreachable" in Fetch as Google contains these two cases). But if that's the case, then Googlebot decided to not try any single resource at all which sounds kind of implausible. // I also can't find a list of error statuses on that help doc. (That's why I had to do a test.)
    – Oskar Skog
    Jul 19, 2017 at 22:38
  • 1
    are your scripts blocked against crawling? robots.txt?
    – Evgeniy
    Mar 1, 2019 at 11:16

4 Answers 4

Reset to default
11

I had 6 instances of "other errors" (4 images and 2 stylesheets) and continually hitting the refresh button didn't help. here's what I think finally fixed it for me:

  1. I switched 2 of the images from relative to absolute paths. that fixed both of those.

  2. I removed the type="text/css" from my css head tags that called the 2 stylesheets. so, they now have only the rel and the href -- like this <link rel="stylesheet" href="https://www.example.com/styles.css">. that fixed the remaining 4 errors. (apparently the remaining 2 images were being called in the stylesheets.)

Improve this answer
6
  • removing text/css fixed it for us. thanks!
    – anonymous-one
    Sep 13, 2019 at 22:54
  • Absolute path fixed this issue, thanks.
    – Borislav Markov
    Jan 22, 2020 at 0:52
  • Absolute paths fixed it for me, too. I wonder if it is the combination with robots.txt and GoogleBot incorrectly processing it?
    – TacoV
    Jun 2, 2020 at 9:51
  • Absolute path for fonts fixes those. After that, the error occurs intermittently with in general a few requests failing. Running on CloudFront CDN, 40 requests, mobile pagespeed 80+.
    – Guido Leenders
    Dec 29, 2020 at 16:28
  • 2
    I doubt that chaning text/css or absolute paths fix anything, it's more likely that you just trigger a re-crawling, which randomly "went better". It looks like the core issue is performance and timing as Google just somtimes rejects some assets (see support thread in @GreatBlakes's answer). I have resources with absolute paths right next to each other and there is no rhyme or reason why some produce "Other error". Also keep in mind that the internal GSC crawler and the manual testing tool are not the same thing.
    – kubi
    Oct 4, 2021 at 8:09
10

According to this support thread, "Other error" may be that Googlebot reached the limit of the number of requests it is willing to make to a server to prevent it from overwhelming the website with requests (AKA "Crawl Budget").

There was not a definitive answer, but that appears to be the response. If this is true, I wish the "error" message was changed to something like "Rate Limit Reached"...

Improve this answer
1
  • I had about 40 small images on my page. Reduced this number by using the "sprite image" method. It fixed the Mobile Friendly Test/Other Error problem.
    – Greg Wittmeyer
    Apr 16, 2021 at 23:50
3

"load 23 page resources" that sounds like your page has a Lot of resources to load.

The mobile tester doesn't like 'heavy' pages.

There are many ways this mimics what it might be like loading it a real mobile device that might have a unstable data connection.

Make the page 'lighter' - loading less 'crud' & the page will load quicker, and hence more 'friendly'.

Improve this answer
4
1

Check your robots.txt to see if it's blocking GoogleBot from any required to load the page.

For example, you could possibly have a CSS script in the <head> that calls a URL that's disallowed in your robots.txt

Improve this answer
4
  • What if I have no robots.txt?
    – Aaron Franke
    Jan 6, 2019 at 2:40
  • 1
    If you have no robots.txt, then nothing is being blocked - so that isn't an issue.
    – xgretsch
    Apr 11, 2019 at 10:36
  • Our website (mrosupply.com) had a separate subdomain (called static.*) served via AWS Cloudfront from a AWS S3 bucket. Digging though the Cloudfront logs it was found that GoogleBot was receiving a 503 LambdaEdge error upon visiting static.mrosupply.com/robots.txt, because the file was inexistent. Simply adding a robots.txt file to the S3 bucket that the subdomain was pointing to, solved the issue for us.
    – symbiotech
    Jan 3 at 14:43
  • the tester will clearly reference the robot.txt file and link to it, if it is the case of the block in my experience. Also sometime your host will pop in a default robots.txt if you delete yours, so always check your url/robots.txt to see if in fact you do have one
    – user13961
    Feb 25 at 18:29

Not the answer you're looking for? Browse other questions tagged or ask your own question.