Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!ucbvax!CIVILGATE.CE.UIUC.EDU!lray From: lray@CIVILGATE.CE.UIUC.EDU (Ray) Newsgroups: comp.sys.apollo Subject: Why is /etc/reboot so unreliable? Message-ID: <8909291722.AA00609@civilgate.ce.uiuc.edu> Date: 29 Sep 89 17:22:11 GMT Sender: daemon@ucbvax.BERKELEY.EDU Organization: The Internet Lines: 33 I've been having considerable problems with /etc/reboot. Over 50% of the time, the node being rebooted fails to come up with one of the following symptoms: 1. The node says the clock is off and waits forever for a y/n. 2. The node freezes prior to going into self tests. 3. The node fails to shut down at all. Condition 1 is the most understandable. If there has been a crash and the hardware clock has been scrambled, then it seems clear a sanity check is in order. Condition 2 could be something to do with the ROM level of the cpu. I have not done enough testing to verify when this happens. Condition 3. is the strangest. It acts like the node is in service mode when I know for a fact it is in normal mode (yes, pressing the reset button works as expected). The problem with all this is that I simply cannot easily get access to the offices where there is a node at all times. This means the node stays down for an unacceptably long period of time while I wait for a key to become available. --------------------------------------------------------------------- Does anyone know how /etc/reboot works, enough to give me some insight as to whats going on here? Leland Ray UIUC - CE Dept. (217) 333-3821