Splunk Search

How can I extract specific data from a complex block of XML?

jtracy
Engager

So I've been reading around and most people point to xpath, but after hours of troubleshooting I can't seem to get it to work in my scenario. I'm trying to extract decision and reasoncode as fields and report on them, but I just can't seem to get Splunk to recognize them in all of the outputs.

So far I have this search working to report all failures (kind of). But I'm hoping for something better.

Sample Search:

host=relevanthost "<c:decision>" "<c:reasonCode>" NOT "<c:decision>ACCEPT</c:decision>"  
|xmlkv |timechart count by c:reasonCode limit=25

Sample Code:

[2016-05-timestampstuff] [HOSTNAME] [NOTIFICATION] [numbers] [oracle.router.stuff] [tid: [ACTIVE].ExecuteThread: '6' for queue: 'weblogic.kernel.Default (self-tuning)'] [userId: <fakeuser>] [ecid: fakenumbersandlettersecid] [oracle.soa.tracking.FlowId: 000000000] [oracle.soa.tracking.InstanceId: 000000000] [oracle.soa.tracking.SCAEntityId: 00000] [FlowId:fakenumbersandletters] [[
 [OSB Tracing] Entering pipeline pair Process Request with message context: 
 [MessageContextImpl  uid="fakeuidlettersandnumbers"
 body="<soap:Body xmlns:soap="http://schemas.fake.url/soap/stuff">
  <c:replyMessage xmlns:c="urn:fake.data.here">
    <c:merchantReferenceCode>0000000000</c:merchantReferenceCode>
    <c:requestID>0000000000000000000000</c:requestID>
    <c:decision>ACCEPT</c:decision>
    <c:reasonCode>100</c:reasonCode>
    <c:requestToken>faketokenlettersandnumbers</c:requestToken>
    <c:purchaseTotals>
      <c:currency>USD</c:currency>
    </c:purchaseTotals>
    <c:ccAuthReply>
      <c:reasonCode>100</c:reasonCode>
      <c:amount>99.99</c:amount>
      <c:authorizationCode>000000</c:authorizationCode>
      <c:avsCode>X</c:avsCode>
      <c:avsCodeRaw>A3</c:avsCodeRaw>
      <c:cvCode>S</c:cvCode>
      <c:cvCodeRaw>S</c:cvCodeRaw>
0 Karma
1 Solution

sundareshr
Legend

If you're only interested in the two fields, you're probably better off using rex. Like this

.... | rex ield=_raw "decision\>(?<decision>[^\<]+)\<" | rex field=_raw "reasonCode\>(?<reasonCode>[^\<]+)\<\/" | where decision!="ACCEPT" | timechart count by reasonCode

View solution in original post

sundareshr
Legend

If you're only interested in the two fields, you're probably better off using rex. Like this

.... | rex ield=_raw "decision\>(?<decision>[^\<]+)\<" | rex field=_raw "reasonCode\>(?<reasonCode>[^\<]+)\<\/" | where decision!="ACCEPT" | timechart count by reasonCode

jtracy
Engager

This worked! Thanks! Would I be able to somehow use this to extrapolate what percentage each value for reasonCode represents in all results with "ACCEPT" as a value for decision in a timechart?

0 Karma

sundareshr
Legend

Try this

.... | rex ield=_raw "decision\>(?<decision>[^\<]+)\<" | rex field=_raw "reasonCode\>(?<reasonCode>[^\<]+)\<\/" | where decision!="ACCEPT" | eventstats count as total by reasonCode | eval p=1/total | bin span=1d _time as time | stats sum(p) as perc by time reasonCode | eval perc=tostring(round(perc, 2), "commas") | eval time=strftime(time, "%x %X") | xyseries time reasonCode perc
0 Karma
Get Updates on the Splunk Community!

Aligning Observability Costs with Business Value: Practical Strategies

 Join us for an engaging Tech Talk on Aligning Observability Costs with Business Value: Practical ...

Mastering Data Pipelines: Unlocking Value with Splunk

 In today's AI-driven world, organizations must balance the challenges of managing the explosion of data with ...

Splunk Up Your Game: Why It's Time to Embrace Python 3.9+ and OpenSSL 3.0

Did you know that for Splunk Enterprise 9.4, Python 3.9 is the default interpreter? This shift is not just a ...