How to export/import selectively data from an indexer to another.
Here is the example for the defaultdb index (the main index) with $SPLUNK_HOME = /opt/splunk and a time period from April 10th 00:00 to April 11th 00:00 GMT (equivalent to 1302393600 to 1302480000 epoch time)
1 - roll the hot buckets to warm on the initial indexer
cd /opt/splunk/bin ./splunk _internal call /data/indexes/defaultdb/roll-hot-buckets -auth admin:changemespecify the correct db name, and password
2- identify the buckets containing data for your time period.
The dates are in epoch time UTC in the filename, in the reverse order. the filename is db_recentevent_oldestevent_bucketuniquenumber. You can use http://www.epochconverter.com/ to check
example : /opt/splunk/var/lib/splunk/defaultdb/db/db_1305913172_1301920239_29/ contains data for the period of to 1301920239 = GMT: Mon, 04 Apr 2011 12:30:39 GMT from 1305913172 = GMT: Fri, 20 May 2011 17:39:32 GMT
3 - export the events for the index and the period you need
usage : exporttool db_directory exportfile [-et <earliest_time_utc>] [-lt <latest_time_utc>] [-csv] [export_search] example : cd /opt/splunk/bin ./splunk cmd exporttool /opt/splunk/var/lib/splunk/defaultdb/db/db_1305913172_1301920239_29/ /myexportpath/export1.csv -et 1302393600 -lt 1302480000 -csvIf needed, you can also add a search as last parameter. Check that an export file was created. Repeat for each buckets containing data of the good period and change the export file. If you want to run the export over all the buckets, use a loop command.
4 - import each file into the new indexer, in the proper destination index
usage : importtool <database_path> <csv_file> example : cd /opt/splunk/bin ./splunk cmd importtool /opt/splunk/var/lib/splunk/defaultdb/db /myexportpath/export1.csv "Successfully imported 71615 events into the bucket. Please ensure this bucket resides in a valid index and restart Splunk to recognize the new events."Restart to have splunk detecting the new data and recalculate the metadata.
example : ./splunk restart .... Perform recovery now? [y/n] y Recovering (across all data)... bucket=opt/splunk/var/lib/splunk/defaultdb/db/db_1306285067_1305920377_54 count mismatch tsidx=2525 source-metadata=2524, repairing... Done
A couple of corrections during import (at least with 4.2.5):
answered 13 Dec '11, 01:23