<?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 Event Types on a Summary Index in Knowledge Management</title>
    <link>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39944#M357</link>
    <description>&lt;P&gt;Team,&lt;/P&gt;

&lt;P&gt;I have a summary index that looks like this:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;&amp;lt;search string&amp;gt; | sistats count by UserAgent
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;I also have a collection of event types that group various UserAgents, such that:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;[ua_iPhone]
UserAgent="iPhone"

[ua_iPad]
UserAgent="iPad"
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;I'd like to query the si and end up with a list of top user agents, sort of like:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;index="summary" search_name="si_useragent" | stats count by UserAgent | eval eventtype=mvfilter(match(eventtype, "ua\_.*")) | top eventtype
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;Is this possible?  Advisable?&lt;/P&gt;

&lt;P&gt;Thanks,
-S.&lt;/P&gt;</description>
    <pubDate>Tue, 31 Aug 2010 01:04:44 GMT</pubDate>
    <dc:creator>srussellnpr</dc:creator>
    <dc:date>2010-08-31T01:04:44Z</dc:date>
    <item>
      <title>Event Types on a Summary Index</title>
      <link>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39944#M357</link>
      <description>&lt;P&gt;Team,&lt;/P&gt;

&lt;P&gt;I have a summary index that looks like this:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;&amp;lt;search string&amp;gt; | sistats count by UserAgent
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;I also have a collection of event types that group various UserAgents, such that:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;[ua_iPhone]
UserAgent="iPhone"

[ua_iPad]
UserAgent="iPad"
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;I'd like to query the si and end up with a list of top user agents, sort of like:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;index="summary" search_name="si_useragent" | stats count by UserAgent | eval eventtype=mvfilter(match(eventtype, "ua\_.*")) | top eventtype
&lt;/CODE&gt;&lt;/PRE&gt;

&lt;P&gt;Is this possible?  Advisable?&lt;/P&gt;

&lt;P&gt;Thanks,
-S.&lt;/P&gt;</description>
      <pubDate>Tue, 31 Aug 2010 01:04:44 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39944#M357</guid>
      <dc:creator>srussellnpr</dc:creator>
      <dc:date>2010-08-31T01:04:44Z</dc:date>
    </item>
    <item>
      <title>Re: Event Types on a Summary Index</title>
      <link>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39945#M358</link>
      <description>&lt;P&gt;Yes, you should be able to do this by manually running the &lt;CODE&gt;typer&lt;/CODE&gt; command after the &lt;CODE&gt;stats count&lt;/CODE&gt;. For example:&lt;/P&gt;

&lt;PRE&gt;&lt;CODE&gt;index="summary" search_name="si_useragent" | stats count by UserAgent | typer | eval eventtype=mvfilter(match(eventtype, "ua_.*")) | top eventtype
&lt;/CODE&gt;&lt;/PRE&gt;</description>
      <pubDate>Tue, 31 Aug 2010 03:36:01 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39945#M358</guid>
      <dc:creator>Stephen_Sorkin</dc:creator>
      <dc:date>2010-08-31T03:36:01Z</dc:date>
    </item>
    <item>
      <title>Re: Event Types on a Summary Index</title>
      <link>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39946#M359</link>
      <description>&lt;P&gt;Stephen has given you an answer. As an aside, I recommend use of a lookup table rather than eventtypes for this use case.&lt;/P&gt;</description>
      <pubDate>Tue, 31 Aug 2010 03:39:50 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39946#M359</guid>
      <dc:creator>gkanapathy</dc:creator>
      <dc:date>2010-08-31T03:39:50Z</dc:date>
    </item>
    <item>
      <title>Re: Event Types on a Summary Index</title>
      <link>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39947#M360</link>
      <description>&lt;P&gt;I'd normally recommend a lookup as well, but my guess is that the actual eventtypes have wildcards, which CSV lookups don't play well with.&lt;/P&gt;</description>
      <pubDate>Tue, 31 Aug 2010 05:24:29 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39947#M360</guid>
      <dc:creator>Stephen_Sorkin</dc:creator>
      <dc:date>2010-08-31T05:24:29Z</dc:date>
    </item>
    <item>
      <title>Re: Event Types on a Summary Index</title>
      <link>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39948#M361</link>
      <description>&lt;P&gt;Ah, precisely!  It looks more like:&lt;/P&gt;

&lt;P&gt;[ua_iphone] UserAgent="&lt;EM&gt;iPhone&lt;/EM&gt;"&lt;/P&gt;

&lt;P&gt;However, I was considering writing a script in python to create a dynamic lookup table, but then I heard about this eventtype approach.&lt;/P&gt;</description>
      <pubDate>Tue, 31 Aug 2010 20:16:14 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Knowledge-Management/Event-Types-on-a-Summary-Index/m-p/39948#M361</guid>
      <dc:creator>srussellnpr</dc:creator>
      <dc:date>2010-08-31T20:16:14Z</dc:date>
    </item>
  </channel>
</rss>

