Hi
I have succesfuly setup a server (ubuntu 6.10se) with slunk (http://www.splunk.com)
Everything works great besides one thing. Event timesteps are wronk. In splunk logs I can see events with tommorows date for example.
I’m remotely logging wireless,info events (connected, disconnected, disconnected, extensive data loss and so on). there are olso strange hours at with data/events are arrived. I atached a picter below.
At first I was thinking that this event are from time after device is restarted and ntp client isn’t synchronized time with local time server. So a write a script which block outgoing 514/udp trafiice after the restart until system clock i synchronized.
ip firewall filter add chain=output protocol=udp dst-port=514 action=drop comment="block syslog";
:while ([system ntp client get status]!=synchronized) do={:delay 1};
ip firewall filter remove [find chain=output comment="block syslog"];
then I clean splunk database. AND i still reciving enevts with bad date and time
I don’t know if it is SPLUNK or MT problem
picture comment: on the top of atached picter there is: Last refreshed, and below are events with tommorows date
looking at logs on MT shows good date and time on remote MT device
digging internet depper i found that it is rather SPLUNK “problem”
SPLUNK at default isn’t recognizing MT syslog timesteps well
if I find a solution I’ll post it here
maybe timestep is coded? i don’t know it yeat. but form now there is a workaround. you must configure splunk to add time/date present on server when remote syslog event arrived.
all you need to do is to setup a network port to 514/udp and set “Set source type” to manual and in “Source Type” type eg. “mikrotik”
then you need to edit $SPLUNK_HOME/etc/bundle/local/props.conf and add this: