Looking at method_thread, it seems like setting the timeout value to infinite(0) and have the method exits with SMF_EXIT_ERR_FATAL will prevent the method from being restarted.
-tony Renaud Manus wrote: > This is RFE 6450661 start method failures should be configurable > > -- Renaud > > ------- Original Message -------- > Subject: [smf-discuss] Method timeouts and retry counts > From: John Levon <john.levon at sun.com> > To: smf-discuss at opensolaris.org > Date: Tue Jun 12 2007 02:21:04 GMT+0200 (MEST) > >> I'm pretty sure this has been discussed before, but I'm in somewhat of a >> quandary: my service would like a reasonably large timeout value, but >> this >> means that if it /does/ go wrong, then whatever value I pick is >> multiplied >> thrice, since a method timeout doesn't cause immediate maintenance. >> >> In my case I'd ideally like to give up straight after my first timeout. >> >> I couldn't find a bug/RFE on this, although there was a related one >> about a >> global timeout on system shutdown. >> >> regards >> john >> _______________________________________________ >> smf-discuss mailing list >> smf-discuss at opensolaris.org > > _______________________________________________ > smf-discuss mailing list > smf-discuss at opensolaris.org
