We've got summary index working great, but we need to back fill in some data from before we started the automated reports that populate the summary index.
Currently we measure conversion rate via this search:
index="ecom" eventName | eventstats dc(sessionId) as totalnumberofsessions | search eventName = orderPlaced | stats dc(sessionId) as numbuyers max(totalnumberofsessions) as total | eval ConvRate = round(numbuyers*100/total,2) | table ConvRate | addinfo | rename info_min_time as _time | table _time ConvRate
In our summary index it's set to -1@d
, @d
. I'm trying to now generate this data on a month by month basis PER day. I tried the following below, but it buckets all the data to the earliest / 1st day of the month
index="ecom" eventName | bucket _time span=1d | eventstats dc(sessionId) as totalnumberofsessions | search eventName = orderPlaced | stats dc(sessionId) as numbuyers max(totalnumberofsessions) as total | eval ConvRate = round(numbuyers*100/total,2) | table ConvRate | addinfo | rename info_min_time as _time | table _time ConvRate
We want it so that it gives the ConvRate for each day based on total # of sessions / sessions that have an eventName called orderPlaced.
Update - I think I solved it?
I think I figured this out - but if someone could confirm it's the right solution or tell me if there is a better way that'd be great.
index="ecom" eventName | bucket _time span=1d | eventstats dc(sessionId) as totalnumberofsessions by _time | search eventName = orderPlaced | stats dc(sessionId) as numbuyers max(totalnumberofsessions) as total by _time | eval ConvRate = round(numbuyers*100/total,2) | table ConvRate _time
Splunk comes with a backfill script for summary indexes - see the docs at http://docs.splunk.com/Documentation/Splunk/6.3.0/Knowledge/Managesummaryindexgapsandoverlaps
Splunk comes with a backfill script for summary indexes - see the docs at http://docs.splunk.com/Documentation/Splunk/6.3.0/Knowledge/Managesummaryindexgapsandoverlaps
Doh! Thanks. Here I've been manually doing it using | addinfo | collect.