home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!paladin.american.edu!news.univie.ac.at!hp4at!mcsun!Germany.EU.net!ira.uka.de!sbusol.rz.uni-sb.de!coli.uni-sb.de!coli.uni-sb.de!usenet
- From: hfi@coli.uni-sb.de (Hannes Fischer)
- Newsgroups: comp.sys.sun.admin
- Subject: Re: /sbin/init burns cpu time?
- Date: 23 Nov 1992 16:00:41 +0100
- Organization: Computational Linguistics Dept., U Saarbruecken
- Lines: 31
- Message-ID: <1eqrmpINNgg9@coli-gate.coli.uni-sb.de>
- References: <1eot3oINN5ac@coli-gate.coli.uni-sb.de> <2B1000F2.11231@ics.uci.edu>
- NNTP-Posting-Host: coli-gate.coli.uni-sb.de
- Keywords: Sun4/25 ELC SUNOS4.1.1B init
-
- sam@john-bigboote.ics.uci.edu (Sam Horrocks) writes:
-
- >In <1eot3oINN5ac@coli-gate.coli.uni-sb.de> hfi@coli.uni-sb.de (Hannes Fischer) writes:
-
- >>Hi everyone,
- >> [init using lots of cpu time]
-
- >Try doing a "trace -p 1". You'll probably see that init is constantly
- >getting a SIGSEGV, recovering and then getting another SIGSEGV. I've
- >seen this happen when making changes to ttytab and sending a SIGHUP to
- >init to get it to re-read the file. Sometimes it'll pick up the
- >changes and sometimes it'll just start SIGSEGV'ing.
-
- >The only fix is to reboot. If you can't do that, renice it +19 until
- >you can reboot.
-
- You hit the nail squarely on the head - I'm enabling/disabling a modem
- getty on ttya out of a cron job (<sribble /etc/ttytab>; kill -HUP 1), and
- init gets into that SIGSEGV loop.
-
- >Sam
- >--
- >Sam Horrocks
- >ICS Department, UC Irvine
- >Email: sam@ics.uci.edu
-
- -Hannes.
- --
- Hannes Fischer, U Saarbruecken, Dept. of Computational Linguistics
- hfi@coli.uni-sb.de (work) - hfi@uhf.saar.sub.org (home)
- #include <std/disclaimer.h>
-