Xref: utzoo comp.unix.questions:31052 comp.unix.misc:1417 comp.unix.sysv386:7815
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!crdgw1!uunet!fciva!dag
From: dag@fciva.FRANKCAP.COM (Daniel A. Graifer)
Newsgroups: comp.unix.questions,comp.unix.misc,comp.unix.sysv386
Subject: Re: What is it? (was Re: Current Run-Level: How can you tell?)
Message-ID: <602@fciva.FRANKCAP.COM>
Date: 5 May 91 16:29:06 GMT
References: <1991Apr23.024433.10460@srwic.UUCP> <1991Apr25.171617.13505@ssd.kodak.com> <230@harem.clydeunix.com>
Reply-To: dag@fciva.UUCP (Daniel A. Graifer)
Organization: Coastal Capital Funding Corp., McLean, VA
Lines: 29


On our Prime EXL's,  which run pretty much Plain Vanilla AT&T SysV 3.1 with
some 3.2 extensions, we have:

	0 - Shutdown System
	1 - Single User Mode
	2 - Normal Multi-User
	3 - Multiuser + remote file sharing
	4 - Undefined by system
	5 - Shutdown and entire ROM diagnostic monitor
	6 - Shutdown and reboot

We've defined to state 4 to be system backup.  A cron job in the wee dawn
hours does a telinit 4 to start the backup process.  We also cleanup the
cron log at that time, since cron isn't running in state 4.

We had to modify some of the scripts in rc2.d:  who -r also contains the
previous run level as well as the current one.  Some of the scripts do
nothing if the previous state isn't 0, eg. RMTMPFILES, and we wanted these
to run coming out of state 4 as well.  (With the backup automated, we've
gone for months without rebooting, so lots of things such as tmp directories
would fill up if we didn't clean them out as part of the nightly backup.)

Dan
-- 
Daniel A. Graifer			Coastal Capital Funding Corp.
Sr. Vice President, Financial Systems	7900 Westpark Dr. Suite A-130
(703)821-3244				McLean, VA  22102
uunet!fciva!dag				fciva.FRANKCAP.COM!dag@uunet.uu.net