Issue with 'Crawl Errors' in Webmaster Tools

[RiceMedia]

Have an issue with a large number of 'Not Found' webpages being listed in Webmaster Tools.  In the 'Detected' column, the dates are recent (May 1st - 15th).  However, looking clicking into the 'Linked From' column, all of the link sources are old, many from 2009-10.

Furthermore, I have checked a large number of the source pages to double check that the links don't still exist, and they don't as I expected.

Firstly, I am concerned that Google thinks there is a vast number of broken links on this site when in fact there is not.

Secondly, why if the errors do not actually exist (and never actually have) do they remain listed in Webmaster Tools, which claims they were found again this month?!

Thirdly, what's the best and quickest way of getting rid of these errors?  Google advises that using the 'URL Removal Tool' will only remove the pages from the Google index, NOT from the crawl errors.  The info is that if they keep getting 404 returns, it will automatically get removed.  Well I don't know how many times they need to get that 404 in order to get rid of a URL and link that haven't existed for 18-24 months?!!

Thanks.

[RiceMedia]

Thanks both for your responses.  It's a strange one and I can only assume that these pages remain in Google's index - I have checked many link sources and found that the links do not exist and therefore haven't done since the page was deleted.  It seems ridicilous that you should have to 301 every page you delete, there are literally 500+ of these phantom links to non-existant URLs and the site is changing all the time.

I have opted to add a 'no index' meta to the 404s and also encourage them to delete from index by adding the pages to the robots.txt file.

Let's see if it works - I'll post on here when I know for sure so other people with the same question can see the outcome.

Thanks again, Damien and Steven.

[sferrino]

Completely agree with Damien. If they don't exist but Webmaster Tools is showing them, 301 them, there has to be a link somewhere on the internet that is causing them to think there is. I would also go through the server logs to see if there is any additional information like a referring page to the non-existent ones.

[krissy-cca]

Hey,

I guess if you've exhausted all other possibilities you can either let them return a 404 and leave them be which will most likely do you no harm or 301 that particular URL to another relevant page on your site.

Make sure they are actually returning a 404 first though via header response check.

DD