Thanks, Chris. I finally got around to checking on this. The debug option helped me discover my real (stupid) error.
The original problem was indeed running without setting 'nodaemon'. I am using supervisor through a buildout recipe - when I thought I set '-n' in the command line of the sub-supervisor it was picking it up as the directory - causing the more severe failure. I instead just set 'nodaemon=true' in the conf file of the sub-supervisor and now all is good. Thanks for the help - and now I know to turn to -e debug. Jim ----- "Chris McDonough" <[email protected]> wrote: > Hmm.. it helps to actually read the whole post. > > Since you say you tried the "-n" flag, I have no idea why this > wouldn't work. > One trick is to use the flags "-n -eDEBUG" on the main supervisor to > see output > from subprocesses in the foreground; this might give you a clue. > > Jim Baack wrote: > > I'm new to supervisor - it's working great for me. I'm trying to run > one or more supervisor instances from a master supervisor - the use > case is one master on a server and sub-supervisors for each user on > the machine allowing them access to only their processes. > > > > The sub-supervisor process is reported as dying each time it is > started by the master - but it is running and the processes it manages > are running too. It tried the -n option on the sub-supervisor - then > it reported as dying and actually didn't run it's processes. > > > > Any suggestions or is what I'm trying not possible/recommended. > Thanks. > > > > Jim > > _______________________________________________ > > Supervisor-users mailing list > > [email protected] > > http://lists.supervisord.org/mailman/listinfo/supervisor-users > > _______________________________________________ Supervisor-users mailing list [email protected] http://lists.supervisord.org/mailman/listinfo/supervisor-users
