Google WMT reports crawl errors on my PDFs

[Joshmc]

In the last 3 months I have continually ran into a problem where many of the PDFs on my site keep coming up in the not found area of webmaster tools. These are valid links and I can load them every time I click them. I was wondering does Google bot have a problem with PDFs? Like is there a way I need to host or optimize them to get them to not error. This doesn't happen for all of my pdfs but a good amount.

Thanks

[tedster]

Do the URLs for those PDF errors contain any odd characters - things that need to be escaped (turned into coded entities) before the browser can send the request?

[g1smd]

What format are the links pointing to those files?

Are they relative, root relative or absolute?

Are there any

../

constructs in those links?

[Joshmc]

Thanks for the quick responses. Their are no special characters just alpha numeric upper and lower case and dashes. I wouldnt think these would be a problem right?

The links are relative (they dont include the domain name, i believe that means relative) and are like this example <a href="/pdf/file-name.pdf">