<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Web Logs Analysis  ( apache ) in Getting Data In</title>
    <link>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13989#M1346</link>
    <description>&lt;P&gt;Hi there.  I'm new to splunk.  Having a bit of trouble getting my head around it ( I know SQL well ) .&lt;/P&gt;

&lt;P&gt;I want to get google clicks ( contains parameter "gclid" ) - then I want to get the ip address for each click and get all the paid clicks between the initial click and the conversion ( uri_location="&lt;EM&gt;success&lt;/EM&gt;"  ) .  Is that a bit hard for splunk to do?  Can you get me start in the right direction? &lt;/P&gt;

&lt;P&gt;Thanks
Martin&lt;/P&gt;</description>
    <pubDate>Fri, 21 May 2010 08:58:53 GMT</pubDate>
    <dc:creator>return2health</dc:creator>
    <dc:date>2010-05-21T08:58:53Z</dc:date>
    <item>
      <title>Web Logs Analysis  ( apache )</title>
      <link>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13989#M1346</link>
      <description>&lt;P&gt;Hi there.  I'm new to splunk.  Having a bit of trouble getting my head around it ( I know SQL well ) .&lt;/P&gt;

&lt;P&gt;I want to get google clicks ( contains parameter "gclid" ) - then I want to get the ip address for each click and get all the paid clicks between the initial click and the conversion ( uri_location="&lt;EM&gt;success&lt;/EM&gt;"  ) .  Is that a bit hard for splunk to do?  Can you get me start in the right direction? &lt;/P&gt;

&lt;P&gt;Thanks
Martin&lt;/P&gt;</description>
      <pubDate>Fri, 21 May 2010 08:58:53 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13989#M1346</guid>
      <dc:creator>return2health</dc:creator>
      <dc:date>2010-05-21T08:58:53Z</dc:date>
    </item>
    <item>
      <title>Re: Web Logs Analysis  ( apache )</title>
      <link>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13990#M1347</link>
      <description>&lt;P&gt;The simple approach will show clicks even after the "success":&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;sourcetype=access_combined (uri="*&amp;amp;gclid=*" OR uri_location="success") | stats list(uri_location) by clientip
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;We can then extend this to exclude the clicks after success:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;sourcetype=access_combined (uri="*&amp;amp;gclid=*" OR uri_location="success") | eval success_time = if(uri_location=="success", _time, null()) | eventstats min(success_time) as success_time by clientip | where isnull(success_time) OR _time &amp;lt;= success_time | stats list(uri_location) by clientip
&lt;/CODE&gt;&lt;/PRE&gt;</description>
      <pubDate>Fri, 21 May 2010 09:54:38 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13990#M1347</guid>
      <dc:creator>Stephen_Sorkin</dc:creator>
      <dc:date>2010-05-21T09:54:38Z</dc:date>
    </item>
    <item>
      <title>Re: Web Logs Analysis  ( apache )</title>
      <link>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13991#M1348</link>
      <description>&lt;P&gt;I also recommend looking at stats, as in Stephen's first example.  &lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;&amp;lt;your search&amp;gt; gclid=* | stats list(uri_location) as multi_uri by clientip | search multi_uri="success" 
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;And the other way I recommend looking at uses the transaction command, which can take little search expressions in the startswith and endswith arguments. &lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;&amp;lt;your search&amp;gt; gclid=* | transaction clientip endswith="uri=success" | search uri=success
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;and the search on the end is to filter out some unwanted fragment transactions that dont end in success.  This may be unnecessary but i think if transaction A ends with success, the transaction command will open another transaction B for that clientip, even though that transaction may never end in success... &lt;/P&gt;

&lt;P&gt;At any rate there are some advantages to using transaction over stats,  one being that you can do maxspan to appropriately count the same clientip more than once over the range.  &lt;/P&gt;

&lt;P&gt;However it's good to get really comfortable with stats because it's generally more powerful and generally has better performance than the alternatives. &lt;/P&gt;</description>
      <pubDate>Fri, 21 May 2010 09:59:51 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Getting-Data-In/Web-Logs-Analysis-apache/m-p/13991#M1348</guid>
      <dc:creator>sideview</dc:creator>
      <dc:date>2010-05-21T09:59:51Z</dc:date>
    </item>
  </channel>
</rss>

