Knowledge Management

How to backfill a summary index with a restricted time for each day?

jyamie
Explorer

I would like to backfill my index up by 2 months. The query however, is time sensitive and requires the day span to be only between 7am-9pm. Currently, my only method is to manually change the earliest and latest times in both the search and the summary index settings to earliest=-1d@d+7h latest=-1d@d+20h, and then to earliest=-2d@d+7h latest=-2d@d+20h, etc. etc.. you can see just how tedious and time-consuming this can become.

Is there any way that I would not have to insert any relative day into my period, to be able to run my overall index search for 30 days with days only involving data between 7am-9pm everyday? if there were an earliest=7h latest=20h kind of deal, that would be great, but I have not found any yet.
Thanks in advance

0 Karma
1 Solution

somesoni2
Revered Legend

You can do something like this

1) Create a scheduled saved search for your query. This query should run daily and summarize data for yesterday
2) For the query, add earliest=-1d@d+7h latest=-1d@d+20h (within the query)
3) For saved search's Start time use -1d@d, for Finish time, use @d (this just to enable scheduling, search'es time range is overridded in the search itself in step 2)
4) For cron use something like this 45 6 * * * (daily once at 6:45 AM)
5) Use the Splunk backfill utility to backfill any duration at once.

Sample command:

Search Time Range= -1d@d to @d
Since, we summarize yesterday's data, to backfill for Jan-01,2015 to Jan-31-2015, the search should run from Jan-02 (to summarize jan-01) to Feb-01 (to summarize Jan-31)

Backfill job Scheduling time
Earliest Time:01/02/2015 00:00
In Epoch: 1420174800  -- to be used as argument et

Latest Time: 02/01/2015 23:59:59.000000
In Epoch: 1422853199   - to be used as argument lt

Backfill Unix Command (after cd $Splunk_home/bin)
./splunk cmd python fill_summary_index.py -app YourAppName -name "YourSearchName" -et 1420174800  -lt 1422853199   -j 1 -dedup true -auth AdminUserName:AdminPasswordName &

View solution in original post

woodcock
Esteemed Legend

You do know about the backfill command, right? You should be able use your standard daily SI-populating search as-is as described here:

http://docs.splunk.com/Documentation/Splunk/6.2.5/Knowledge/Managesummaryindexgapsandoverlaps

0 Karma

jyamie
Explorer

actually i didn't know about the backfill command. bit of a splunk noob here. thanks!

0 Karma

somesoni2
Revered Legend

You can do something like this

1) Create a scheduled saved search for your query. This query should run daily and summarize data for yesterday
2) For the query, add earliest=-1d@d+7h latest=-1d@d+20h (within the query)
3) For saved search's Start time use -1d@d, for Finish time, use @d (this just to enable scheduling, search'es time range is overridded in the search itself in step 2)
4) For cron use something like this 45 6 * * * (daily once at 6:45 AM)
5) Use the Splunk backfill utility to backfill any duration at once.

Sample command:

Search Time Range= -1d@d to @d
Since, we summarize yesterday's data, to backfill for Jan-01,2015 to Jan-31-2015, the search should run from Jan-02 (to summarize jan-01) to Feb-01 (to summarize Jan-31)

Backfill job Scheduling time
Earliest Time:01/02/2015 00:00
In Epoch: 1420174800  -- to be used as argument et

Latest Time: 02/01/2015 23:59:59.000000
In Epoch: 1422853199   - to be used as argument lt

Backfill Unix Command (after cd $Splunk_home/bin)
./splunk cmd python fill_summary_index.py -app YourAppName -name "YourSearchName" -et 1420174800  -lt 1422853199   -j 1 -dedup true -auth AdminUserName:AdminPasswordName &
Get Updates on the Splunk Community!

.conf24 | Registration Open!

Hello, hello! I come bearing good news: Registration for .conf24 is now open!   conf is Splunk’s rad annual ...

ICYMI - Check out the latest releases of Splunk Edge Processor

Splunk is pleased to announce the latest enhancements to Splunk Edge Processor.  HEC Receiver authorization ...

Introducing the 2024 SplunkTrust!

Hello, Splunk Community! We are beyond thrilled to announce our newest group of SplunkTrust members!  The ...