Date: Tue, 17 Jan 2012 19:43:46 +1100 From: Aristedes Maniatis <ari@ish.com.au> To: Jeremy Chadwick <freebsd@jdc.parodius.com> Cc: freebsd-stable <freebsd-stable@freebsd.org> Subject: Re: dumpdev default Message-ID: <4F1534C2.4080803@ish.com.au> In-Reply-To: <20120117081040.GA18825@icarus.home.lan> References: <4F15253E.7070404@ish.com.au> <20120117081040.GA18825@icarus.home.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
On 17/01/12 7:10 PM, Jeremy Chadwick wrote: > On Tue, Jan 17, 2012 at 06:37:34PM +1100, Aristedes Maniatis wrote: >> The manual states that dumpdev "AUTO is the default as of FreeBSD 6.0" [1] >> >> However: >> >> # uname -a >> FreeBSD xxxxxx 9.0-RELEASE FreeBSD 9.0-RELEASE #0: Tue Jan 3 07:46:30 UTC 2012 root@farrell.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC amd64 >> >> # grep dumpdev /etc/defaults/rc.conf >> dumpdev="NO" # Device to crashdump to (device name, AUTO, or NO). >> savecore_flags="" # Used if dumpdev is enabled above, and present. >> >> >> It looks like NO is still the default. Is there a reason why this should not be turned on even for production machines? I haven't read about any side effects, but it seems to be off by default for some reason. > > The Handbook is incorrect, and I filed a PR for this matter last year > (PR 159650): > > http://lists.freebsd.org/pipermail/freebsd-doc/2011-August/018654.html > > Worth reading: > > http://lists.freebsd.org/pipermail/freebsd-stable/2011-August/063541.html > http://lists.freebsd.org/pipermail/freebsd-stable/2011-August/063542.html > http://lists.freebsd.org/pipermail/freebsd-stable/2011-August/063543.html > > And the entire thread: > > http://lists.freebsd.org/pipermail/freebsd-stable/2011-August/063535.html Ahh!!! Someone give Jeremy a commit bit for the FreeBSD documentation already! The commit you reference by mnag has only the following log: -------------- - Change dumpdev default to "NO". Only HEAD is set to "AUTO" Discussed with: re Approved by: re (scottl) -------------- Not enough to know whether dumpdev is now considered a dangerous feature for production servers. Only scottl or mnag may know the answer. I've also found another bug. man dumpon(8) shows: The dumpon utility will refuse to enable a dump device which is smaller than the total amount of physical memory as reported by the hw.physmem sysctl(8) variable. However, I have found that # dumpon -v /dev/ad4s1b kernel dumps on /dev/ad4s1b # sysctl hw.physmem hw.physmem: 25744310272 # swapinfo Device 1K-blocks Used Avail Capacity /dev/ad4s1b 8388608 0 8388608 0% So, 24Gb RAM, 8Gb swap device. No complaints from dumpon. Either it is silently failing (poor response from the app) or it is incorrectly setting the dump device to something too small. Ari -- --------------------------> Aristedes Maniatis ish http://www.ish.com.au Level 1, 30 Wilson Street Newtown 2042 Australia phone +61 2 9550 5001 fax +61 2 9550 4001 GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4F1534C2.4080803>