How long will Google take to recognize 301?

From non-www to www.

[Blade3]

Hi

I redirected my site from non-www to www,( only the root ) on Feb 8,2006. Still I find different pages for site:domain.com and site:www.domain.com.

Is it ok if I redirect only the root or I have to redirect each and every page of my site?

When can I expect Google to recognize the redirection?

Note: All non-www pages are appearing as supplemental results.

[tedster]

In the past, I've seen times of 4-6 weeks to sort this out. At this moment, the rolling out of the Big Daddy infrastructure may change this. First, Google resources are quite involved in Big Daddy. But more important, Google hopes that the new infrastructure will give them a better handle on canonical issues altogether, including the no-www thing.

No matter which of these factors you look at, 10 days is too quick to hope for a change. Hang in there!

[arnarn]

I thought BD was still trying to sort this out and that we may be talking years?

[av8erab]

I have been trying to fix this issue on my site. Are you saying that there may be no need to fix this in the future?

Thanks

[Blade3]

Tedster

>>Google hopes that the new infrastructure will give them a better handle on canonical issues altogether, including the no-www thing.

Do you think Google with the new infrastructure will consider www.domain.com and domain.com as one and the same?

Or will it respond quickly to these type of redirections?

"BigDaddy what's there in your 'bag'?" ;-)

[watercrazed]

I implemented 301 redirects on every page of non.www in november the supplemental results are still there.

[g1smd]

For a 301 redirect from non-www to www, Google will start to correctly index the www pages within just a week or two. The number will increase, and more will gain a title and description in the SERPs (as long as every page has a unique title and description that relates to the content on that page). It will take a few months for the www pages to be fully indexed.

It will take Google at least several months to drop the unwanted non-www versions of each URL. They will mostly turn into URL-only entries first, before dropping out.

A number of entries will turn into (or will re-appear as, several months after the initial drop) Supplemental Results and Google will hang on to those for two or three years. You cannot control that part at all.

[Dayo_UK]

Google has not sorted it out correctly for my site after almost a year of a 301 redirect.

Sometimes it seems to have fixed it but Google always seems to have a back up/base copy that it can revert to - the lastest PR update has split a site of mine again - the 301 was done in March last year.

Roll on Big Daddy.

[steveb]

"When can I expect Google to recognize the redirection?
Note: All non-www pages are appearing as supplemental results."

Never. Currently Google never recognizes the redirection of a supplemental result.

[g1smd]

You say you have only redirected the root index page.

You need to redirect every page of the site from non-www to www using a 301 redirect. This can be done with just 3 lines of code in your .htaccess file.

[pincher34]

Don't hold your breath. I used .htaccess to make all non-www's redirect to the www version over a year ago. Google still proudly displays the non-www pages. I used a Google sitemap, but that made no difference either. They never went to URL only (no title or description), though they do show as supplemental. Even some pages that have been added to the site after the .htaccess file was in place, have shown up as non-www's.

Edited for spelling.

[g1smd]

>> Even some pages that have been added to the site after the .htaccess file was in place, have shown up as non-www's. <<

Hmm, Please get yourself a copy of WebBug and check that those URLs really do return the "301" response code.

If the non-www URL has been indexed for new pages, then it is my guess that those URLs actually return a 302 or 200, and not the required 301 status. Please check it.

[phish]

The lentgth of time has no bearing. The question is how long will it take them to get your 301 right. They still dont understand them. Even Big Daddy (cough cough) is a sad attempt at the moment. JMHO

[tedster]

In my experience with the no-www issue, I never relied solely on the 301 redirect ro fix an already existing problem. I wanted to maximize the cues that Google was getting, so in addition to the domain-wide 301 for all urls, I added the <base href=""> element to every head section, and I also scrutinized every url (on-site and off-site) for that lousy no-www linking.

So when I say 4 to 6 weeks for Google to straighten this out, I always was using a combined approach. Also, I've only worked with clients on an IIS server to resolve this problem and the IIS server has a very direct way to create the 301 redirect from within "Internet Services Manager". Fortunately, none of my Apache clients have had a no-www problem. But that also means I haven't been hands-on with htaccess solutions, and from what I hear, there are pitfalls and potential misconfigurations here, too.

One thing I'm sure of is this -- if the problem is there, then extreme technical care is essential for getting the hole plugged up. A 301 must really and always be a 301, all the urls in the base tag must be precise, and link urls throughout the site must all be accurate.

[theBear]

If I can believe the site command I have exactly one remaining page out of several thousand that hasn't had the 301 recognised.

We had a couple thousand pages that got spread out over multiple vanity domains both with and without www and IP address to round out the field, we at one time had 5 copies of several hundred pages,

Our 301's went in in March of last year.

We also helped Google crawl links so it picked up the 301s.

Things were messy for a long while.

[Blade3]

After 301 is done, when I searched for site:www.domain.com there are two results www.domain.com and www.domain.com/index.html with different Titles.

www.domain.com shows changed title and www.domain.com/index.html with old title.

Need your expert advice. Shall I redirect www.domain.com/index.html to www.domain.com/?

[jdMorgan]

Blade3,

> Do you think Google with the new infrastructure will consider www.domain.com and domain.com as one and the same?

No, they'll never completely do this, because it would be an error. Those are two different domains, and the only way Google can tell that you want them 'to be the same' is that they see identical content most of the time when comparing pages between the two. When doing this comparison, they have to allow for the fact that you might change a page after they index it on one doamin and before they index it on the other. That's one reason that sites that rely on Google to 'figure it out' are so fragile.

As to redirecting "/index.html" to "/", yes, you should if they resolve to the same page. But it's tricky. The required method depends on your server.

Jim

[pincher34]

g1smd - yes, the non-www's all return 301's on a server header check. I also have the base href tag, and I use absolute links, not relative. My PR is 5, and site's been online since 2001, fwiw.

[Strider]

fighting with that as well...
Dear JDMorgan this doesnt seem to work together:

RewriteCond %{HTTP_HOST}!www.sitename.eu.com
RewriteRule (.*) [sitename.eu.com...] [R=301,L]
RewriteCond %{REQUEST_URI} ^/index\.html
RewriteRule ^$ / [R=301, L]

what am I doing wrong please?

[jdMorgan]

I warned you it was tricky: :)

RewriteCond %{THE_REQUEST} ^[A-Z]{3,9}\ /index\.html\ HTTP/
RewriteRule ^index\.html$ http://www.example.com/ [R=301,L]

Jim

[g1smd]

How do you modify that to work with all index.html files in all subdirectories?

I still don't understand what the ^[A-Z]{3,9}\ bit does either.

[Strider]

Thanks Jim :)

I am also trying to understand what this bit does.

You're the best Jim :)

[jdMorgan]

One way to catch all subdirectory index files:

RewriteCond %{THE_REQUEST} ^[A-Z]{3,9}\ /(([^/]+/)*)index\.html\ HTTP/
RewriteRule index\.html$ http://www.example.com/%1 [R=301,L]

The [A-Z]{3,9} pattern is intended to match the GET, POST, PROPFIND, etc. HTTP methods. {THE_REQUEST} is the exact request header sent by the browser, for example:

GET /forum/index.php?t=1302 HTTP/1.1

This is exactly as a request would appear in your raw server access log.

The shortest valid HTTP method is three characters long, and the longest is 9.

Jim