-the old curator does not support ES 5.x
-include curator 5.1.1 and pin version to exactly that to avoid surprises with disruptive updates
-configs reside in /etc/tpot/curator
-will be started daily through /etc/crontab
-by default all logstash indices older than 90 days will be deleted
if persistence is enabled, log files, downloads, binaries, etc. will be compressed and rotated
each start / stop of the t-pot service will account for a full rotation cycle if files are not empty
basically the rotation will recycle logs after 30 days, unless the service is stopped / started manually which will cause for a shorter period
do not forward tcp connections to or from 127.0.0.1 to NFQ (fixes strange netdata behaviour)
run netdata on network mode host again (update compose files) including host metrics
use docker-compose from pypi with support for 2.1 compose file version
logstash, kibana, head & netdata are now depending on a healthy elasticsearch container before starting
remove alerta-cli
tweak installer
update dps.sh
adjust docker-compose related tpot configs for dionaea (stdin_open: true)
adjust tpot.service (suritcata / p0f prep) to be aware of a situation without local network route ( Fixes#99 )
