Date: Mon, 30 Oct 1995 09:19:31 +0100 (MET) From: J Wunsch <j@uriah.heep.sax.de> To: andreas@knobel.gun.de (Andreas Klemm) Cc: jkh@time.cdrom.com, hackers@freebsd.org Subject: Re: several diffs for dump(8) to display write throughput and such...c Message-ID: <199510300819.JAA04834@uriah.heep.sax.de> In-Reply-To: <199510291853.TAA04405@knobel.gun.de> from "Andreas Klemm" at Oct 29, 95 07:53:46 pm
next in thread | previous in thread | raw e-mail | index | archive | help
As Andreas Klemm wrote: > > > > >>DUMP: DUMP: finished in 1451 seconds, throughput 474 KBytes/sec > > > > Neat! I like it. However, we're already in feature freeze for 2.1 so > > this would definitely be -current fodder only, I'm afraid! > > Hi Jordan, well, that's fine. So you take the part to fiddle it in > into current, please ?! I've already wrote this to Andreas in private mail: o The patch could cause a division by 0 under rare circumstances, it divides by a time difference without first checking for 0. The probably best way is to add a `1' there. It won't change the result very significantly for a long-term tape backup, but avoids the problem. o I vote for /dev/rst0. This is consistent with the default tape device of tar(1), and while the non-rewind device is more appropriate for people with large tape drives (DAT, QIC 1 GB or above), the rewind device is a better default for small tape drives like the very common QIC-150 ones, where you sometimes even need more than one cartridge for the entire dump, so the tape _must_ be rewound by dump(8). If you forgot this, you have to ^Z/mt rewind/fg! -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510300819.JAA04834>