I'm running free version 4.2.3. I have an email archive that I'm pre-processing the data to allow for key=value settings. I need to have one field that is the string content including multiple lines and >> responses and other characters that I need to stuff into a field. I control the data format, so I've put an #EOR (end of record) string into the data.
I've been through the documentation for multi-line events here: http://docs.splunk.com/Documentation/Splunk/4.2.3/Data/Indexmulti-lineevents
There wasn't a copy of C:/program files/splunk/etc/system/local/props.conf, so I made one.
I tried to use TRUNCATE = 0 but it stuck everything in one single record. Also LINE_BREAKER = "#DCS_EOR"
Please advise on if I'm editing the right file, file location, values, have the right regex for the line breaker, etc.
Please see two records fo data below:
m_id="27001",t_id="26044",t_name="Simpsons: BT Raw supply",p_id="1434",p_firstname="Tabish",p_lastname="Hasan",p_fullname="Tabish Hasan",m_idparent="0",m_type="email",fileid="3",from_firstname="Tabish",from_lastname="Hasan",from_fullname="Tabish Hasan",m_createdate="2007-07-26","to"="torrents","attachments"="multipart/alternative:/opt/dcsemap/visualizer/database/repository/f_file/00/00/00/03/contents/mediadefender@codebrilliant.com/Inbox/attachment/mdexch01.mediadefender.com/F55A72EF782B43428EBF605551914B20E8C474/attachment1.1,text/html:/opt/dcsemap/visualizer/database/repository/f_file/00/00/00/03/contents/mediadefender@codebrilliant.com/Inbox/attachment/mdexch01.mediadefender.com/F55A72EF782B43428EBF605551914B20E8C474/attachment1.1.2",m_content="FYI...for those that look at BT raw supply through the munger.
The BT Raw Supply for The Simpsons Movie is inaccurate. It's picking up
our decoys and counting it as supply. There's not really that much
non-MD supply out there for it yet.
Gerald and Arthur are looking into it.
-TH
"
#DCS_EOR
m_id="27002",t_id="26045",t_name="Status Change - [PM !GOR-850395]: Mark as Single (UMG)",p_id="1432",p_firstname="Gilberto",p_lastname="Vargas",p_fullname="Gilberto Vargas",m_idparent="0",m_type="email",fileid="3",from_firstname="Gilberto",from_lastname="Vargas",from_fullname="Gilberto Vargas",m_createdate="2007-09-05","to"="Jay Mairs","attachments"="multipart/alternative:/opt/dcsemap/visualizer/database/repository/f_file/00/00/00/03/contents/mediadefender@codebrilliant.com/Inbox/attachment/support.mediadefender.com/jnwzva.6mzu7b/attachment1.1,text/html:/opt/dcsemap/visualizer/database/repository/f_file/00/00/00/03/contents/mediadefender@codebrilliant.com/Inbox/attachment/support.mediadefender.com/jnwzva.6mzu7b/attachment1.1.2",m_content="Ticket Status: Mark as Single (UMG)
New Status: Closed
Old Status: On Hold
Ticket Details
===================
Ticket ID: GOR-850395
Department: QA
Priority: Low
Status: Closed
Link:http://support.mediadefender.com/support/staff/index.php?_m=tickets&_a=viewticket&ticketid=14634"
#DCS_EOR
You are using a file that should work and there are many ways to do it. Try this way:
[MySourceType]
LINE_BREAKER = (#DCS_EOR)
SHOULD_LINEMERGE = false
The docs are here:
http://docs.splunk.com/Documentation/Splunk/latest/Data/Indexmulti-lineevents
The key statement I suspect you missed is this:
"The regex must contain a capturing group -- a pair of parentheses that defines an identified subcomponent of the match."