HTML suggestions - Duplicate title tags report

[Vrindavan]

HTML suggestions - Google Webmaster Tools
Duplicate title tags

Today i found a lot of results like the following in the report, no clue what are they

Index of /animals
/animals/?DA
/animals/?MA
/animals/?MD
/animals/?SA
/animals/?SD

Index of /marine
/marine/?DA
/marine/?MA
/marine/?SD

Index of /election
/election/?DA
/election/?MD
/election/?ND
/election/?SA

[penders]

...no clue what are they

Presumably these do relate in some way to the pages on your site?

It looks like you might be accessing/linking to the same page with different query strings, which may or may not alter the content on the page? However, the page title remains constant. It looks like you need a rel="canonical" link referring to the base URL and thus removing the 'duplicate' pages?

[phranque]

the preferred solution is usually a 301 redirect to the canonical url.

have you checked your server access logs to determine who is requesting these non-canonical urls and which urls are referring these requests?

[Vrindavan]

clicking on any links above will not see any webpage.

in those sub folders, no index file present, only other sub folders.

i never link to them like this and certainly not using " ? "

All site are static pages only.

If the problem is found from Google (HTML suggestions - Duplicate title tags report), the cause is google bot rather than other bots ?

i doubt my site will be link from other sites in those linking format.

[koan]

It may not be your fault at all, just other sites linking to you, adding parameters for their own tracking needs. Some of my sites in Webmaster Tool have duplicate title problems because Google found links that always add &referer=sphere_search at the end. It's annoying and I haven't figured out how to fix it without doing some PHP 301 redirect for every possible url parameters that shouldn't be there. The canonical approach is interesting.

[phranque]

clicking on any links above will not see any webpage.

googlebot's request apparently got enough of a document to extract a title element.
you should be responding to these requests with either a 301, 404 or 410 status code.

in those sub folders, no index file present, only other sub folders.

you should be responding to these requests with either a 403 or 404 status code or allow the server to create a directory index on the fly if appropriate.

i never link to them like this and certainly not using " ? "

this is ok but it doesn't solve the problem caused by google discovering these urls elsewhere.

If the problem is found from Google (HTML suggestions - Duplicate title tags report), the cause is google bot rather than other bots ?

the problem may have been discovered and reported by google but certainly wasn't caused by googlebot.

i doubt my site will be link from other sites in those linking format.

this is actually the most likely scenario but you can only be sure if you check your server access logs.
you can also find the traffic sources of your content if you have google analytics on your site.

[Vrindavan]

Is that google analytics cannot count traffic landing on URLs ( 404 Not Found error message) ?

[lucy24]

have you checked your server access logs to determine who is requesting these non-canonical urls and which urls are referring these requests?

I think GWT itself has the same information. It does for vanilla 404's: there's a column that says "linked from so-and-so-many pages" and it will give you a list. Now, if the linking pages themselves no longer exist-- or never existed in the first place-- then you've got some hunting to do.

Come to think of it, are the nonexistent pages listed as both "duplicate title" and 404? I don't think anything in GWT is mutually exclusive.

[phranque]

the problem described in the OP is "Duplicate title tags" which is in the "HTML suggestions" section of Google Webmaster Tools.
each title entry is linked to the list of urls on your site where that duplicate title was found.

if a requested/referred url is 404 then there is no title tag/no duplicate.

[phranque]

Is that google analytics cannot count traffic landing on URLs ( 404 Not Found error message) ?

i don't understand your statement/question.
these urls are not 404 - if they were there wouldn't be a duplicate title problem.

[lucy24]

these urls are not 404 - if they were there wouldn't be a duplicate title problem.

They could be both, with different Discovery Dates.

clicking on any links above will not see any webpage.

If the pages don't exist, and clicking on the links doesn't lead to anything, where are the duplicate titles coming from? It seems a little unfair to cry Duplicate Title when the title in question is "404--Page Not Found" ;)

I haven't had a "Duplicate Title" in a while. Is there a "date discovered" column as in the Crawl Errors? How old are the nonexistent pages?

[phranque]

They could be both, with different Discovery Dates.

that's possible, but i would guess that one would cancel out the other upon discovery or shortly thereafter.

If the pages don't exist, and clicking on the links doesn't lead to anything, where are the duplicate titles coming from? It seems a little unfair to cry Duplicate Title when the title in question is "404--Page Not Found"

it's more than fair if that's the title of your 404 error page and your server is responding with a "soft 404" ("Not Found" error page and a 200 OK status code response or 301/302 redirect to a 200 OK "error page")
=8)

and no discovery dates for HTML suggestions in GWT.

[potshr]

this could be a case where you loaded a wrong header or other include file of some other website by mistake and in that while the bot indexed it. it will now show you those pages where by mistake your website wanted to take, the pages actually donot exist on this domain.
did that happen?

[edited by: engine at 10:24 am (utc) on Dec 19, 2011]
[edit reason] self promo [/edit]

[Dijkgraaf]

Those /?DA etc. look like corrupted directory sorting options for Apache.
?M=D sorts the directory by last modified date, descending.
?D=A sorts the directory by file description, ascending
?S=A sorts the directory by file size, ascending
etc.

And by what you've said so far, I'd say Google is getting a directory listing.
So you either
1) need to change the settings on your sever so it disallows directory listing.
2) Put redirect rules in place.
or 3) or put an index page in each directory with a either with canonical url or that issues a http 410