I would start by putting "set -x" at the top of the script and looking at the log file to see why it's exiting.
o peltamoto wrote: > We have a locally developed syslog-ng pkg that works perfectly on 200+ > systems of various architectures and kernel revs. The SMF service created > with the package has been great so far on all of these systems, however there > is one system that just doesn't work with SMF. Whenever it tries to start it > goes into maintenance mode: > > svc:/system/system-log-ng:default (system log) > State: maintenance since Wed Oct 22 19:35:32 2008 > Reason: Restarting too quickly. > See: http://sun.com/msg/SMF-8000-L5 > See: syslogd(1M) > See: /var/svc/log/system-system-log-ng:default.log > Impact: This service is not running. > > [ Oct 22 19:35:32 Executing start method ("/lib/svc/method/system-log-ng") ] > syslog-ng service starting. > [ Oct 22 19:35:32 Method "start" exited with status 0 ] > [ Oct 22 19:35:32 Stopping because all processes in service exited. ] > [ Oct 22 19:35:32 Executing stop method (:kill) ] > [ Oct 22 19:35:32 Restarting too quickly, changing state to maintenance ] > > Ok, so it seems like it may be a problem with the start method - but I can > run that manually just fine: > > # ps -ef|grep syslog-ng > root 1614 29276 0 22:07:07 pts/9 0:00 grep syslog-ng > # /lib/svc/method/system-log-ng > syslog-ng service starting. > # ps -ef|grep syslog-ng > root 1622 29276 0 22:07:18 pts/9 0:00 grep syslog-ng > root 1620 1 0 22:07:16 ? 0:00 /usr/local/sbin/syslog-ng > > > and like I said before - this same package runs fine on many other systems. > How can I further debug this service start? > -- > This message posted from opensolaris.org > _______________________________________________ > smf-discuss mailing list > smf-discuss at opensolaris.org > >
