Add cronjob to conan instance to check alive-ness

[partouf]

This sometimes crashes because of unknown reasons and I have to manually restart it.
Would be nice to have some kind of automated recovery attempt

Status for service ce-conan should be running

(Should not be done through AWS healtcheck, the instance cannot be scaled, restart is possible but takes too long and there's really no need to do that. /sbin/service ce-conan start should be good enough or ce conan restart from admin node)