Date: Tue, 18 Dec 2012 14:08:17 -0500 From: John Baldwin <jhb@freebsd.org> To: Pawel Jakub Dawidek <pjd@freebsd.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r244320 - head/sbin/savecore Message-ID: <201212181408.18179.jhb@freebsd.org> In-Reply-To: <201212162306.qBGN6CZR069280@svn.freebsd.org> References: <201212162306.qBGN6CZR069280@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday, December 16, 2012 6:06:12 pm Pawel Jakub Dawidek wrote: > Author: pjd > Date: Sun Dec 16 23:06:12 2012 > New Revision: 244320 > URL: http://svnweb.freebsd.org/changeset/base/244320 > > Log: > Implement -m option to savecore(8) that allows to limit number of kernel > dumps stored. Once the limit is reached it restarts from 0. Why restart at zero? The old behavior is that if you rm'd /var/crash/vmcore.0 then new dumps would just get increasing numbers. That seems fine (just delete the "oldest" core dumps if you are out of room). I guess the restarting lets you be lazy and avoid finding the "oldest" dump via a glob, but: The real feature request was not a limit on the number of core dumps IIRC, but a way to specify a minimum amount of free space in the partition and to delete tne oldest dump if the new dump would put the partition over the limit. To do this you have to be able to find the "oldest" dump, so if you solve this you no longer have to rely on rotating names (and no longer need a 'last' link). -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201212181408.18179.jhb>