For below list of data stored in a files, the first line is the fields name and exact data is started on 2nd line.
However, after indexed, it merged to a single events instead of multiple events.
How could the events break down into multiple events? Thanks.
"Number" "Date" "Time" "Interface" "Origin" "Type" "Action" "Service" "Source Port" "Source" "Destination" "Protocol" "Rule" "Rule Name" "Current Rule Number" "User" "Information" "Product" "Source Machine Name" "Source User Name"
"2495" "5Nov2013" "0:00:10" "Internal" "uabcabc" "Log" "Drop" "microsoft-ds" "1164" "123.123.123.123" "www.abcabc.com" "tcp" "52" "" "" "" "" "VPN-1 Power/UTM" "" ""
"2523" "5Nov2013" "0:00:13" "Internal" "uabcabc" "Log" "Drop" "http" "50895" "123.123.123.123" "www.abcabc.com" "tcp" "14" "" "" "" "" "IPS Software Blade" "" ""
"2524" "5Nov2013" "0:00:13" "Internal" "uabcabc" "Log" "Drop" "http" "50898" "123.123.123.123" "www.abcabc.com" "tcp" "14" "" "" "" "" "IPS Software Blade" "" ""
Check splunkd.log to see what it has to say about the time extraction. The timestamp processor is usually pretty good about letting you know of any errors.
tried this props.conf, but didn't work 😞
Any hints?
props.conf
TIME_PREFIX = "\d*"\s"
MAX_TIMESTAMP_LOOKAHEAD = 21
TIME_FORMAT = %d%b%Y" "%H:%M:%S
This happens when Splunk doesn't find a valid timestamp in the message - by default Splunk will break into a new event when it finds a new line with a valid timestamp on it. You need to tell Splunk how to identify the timestamp by setting a valid TIME_FORMAT
string. See more info here: http://docs.splunk.com/Documentation/Splunk/latest/Data/Configuretimestamprecognition