Blocking Spiders

How to block spiders from crawling your site

[Brian0275]

We are getting ready to implement a feature on our site that would create massive duplicate content. It is good for users, bad for spiders. We have weighed several options about how to avoid this.

1) rel="canonical" - I think this is not the best option because it will still allow spiders to use up our resources

2) block all major bot User Agents from seeing this feature - Would require updating, would never be able to block all bots.

3) Meta noindex, nofollow, noarchive - Could be a possible solution

What do you guys suggest/had the best luck with?

[penders]

4) Block known spider IPs? (Is there a list of known spider IPs?) ... would perhaps need updating like the User Agents.

5) Robots.txt?

[tangor]

Work smart, not hard... WHITELIST the bots allowed and disallow all the rest. These days that list of who's invited in is MUCH SMALLER than the other way around.

[g1smd]

Bad bots ignore robots.txt so you need .htaccess rules to shut the door in their face.

Good bots obey robots.txt so you can keep them out of various bits of the site quite easily if those URLs have an easy to recognise pattern or feature in them.

[tangor]

Thanks for reminder that I left out a bit, g1smd... .htaccess, of course, is how it is done. As for robots.txt, same applies: Whitelist what you allow and disallow all the rest... then deal with those unruly bots which do not play nice. Mea cupla for leaving that out as we do say this many, many times in many, many threads... my bad...!

[g1smd]

It's a multi-faceted approach. Once you have a month of raw site logs you will have details for much more than 90% of the bots that might access your site.

After that, you'll just need a minor tweak to the rules now and again.