How to learn{find out} on what searches find my site in search machines?

There are two possible{probable} ways most precisely to learn{find out} search phrases on which find a site:

1.         The most exact way - to analyse raw-broad gullies of the web-server. Each web-server creates such broad gullies and saves in them all facts of the reference{manipulation} of users to the information on a site. Usually, values of a field are stored{kept} in these broad gullies "Referer", transferred{handed} by a browser (so-called referery). In this field the address of page is passed, under the link whence the user has passed on the current page. Analyzing addresses in this field it is possible to find without effort not only concrete search phrases, but also pages of results of search where links to pages of a site cost{stand};

2.         Some counters, such as Rejting@Mail.ru, HotLog, etc. save and show the information on search phrases on which users come on a site. As a matter of fact they use the same contents of a field "Referer", but possess smaller accuracy because of technical features of job. Also there are analytical methods which do not give an exact picture. But allow, not having access to broad gullies of a site and to reports of counters to receive the rough information.