On Mon, May 16, 2016 at 01:12:08PM +0200, Lutz Mader wrote: > Hi all folks, > I start using monit some months ago and got a problem all the time a > "Process" or "Program" was stopped via monit and became "Not monitored" > (the process was gone), but started by someone else manually later > (started outside from monit).
Apart from your training issue of staff not using correct daemon start procedures, it sounds like you'd get more mileage from configuration management (ansible/chef/puppet/salt/etc.) for this issue. That way every configuration management application would return the daemon to its correct state, and could also update your monit configuration (including monit reload) besides. > This is sometimes a problem, because independent of a process is running > or stopped, monit gives the information "Not monitored" all the time and > there is no way to stop this "Zombie“ via monit. > The service monitoring mode is active all the, this is the default. > > My question, how to stop a process via monit permanently. > How to identify such a manually startet process and how to get the right > state for this processes. > > Today my understanding of "Not monitored" is, monit ignore the "Process" > or "Program" (the "status" and "monitoring status" is set to "Not > monitored"), > but this is not something like a "Stopped" status. > > Any suggestion how to handle this, > Lutz > > p.s. > I start using monit in january this year with some AIX, Linux systems to > migrate the application startup, stopping and monitoring to a more > useful tool. > In the past we, I use lots of application specific start, stop, monitor > scripts, but this doesn't fit my understanding of the way how to do this. > > I start using monit to get a useful tool to do all the systems > housekeeping in a easy way with a uniform interface to handle all the > actions. > > -- > To unsubscribe: > https://lists.nongnu.org/mailman/listinfo/monit-general -- To unsubscribe: https://lists.nongnu.org/mailman/listinfo/monit-general
