Date: Thu, 7 Oct 2004 06:50:50 -0400 From: "Kevin A. Pieckiel" <pieckiel@sdf.lonestar.org> To: Dan Nelson <dnelson@allantgroup.com> Cc: freebsd-questions@freebsd.org Subject: Re: Tape xfers topping out at 10K per transaction? Message-ID: <20041007105050.GA7349@SDF.LONESTAR.ORG> In-Reply-To: <20041007032416.GK3848@dan.emsphone.com> References: <20041007025138.GA19296@SDF.LONESTAR.ORG> <20041007032416.GK3848@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Oct 06, 2004 at 10:24:16PM -0500, Dan Nelson wrote: > In the last episode (Oct 06), Kevin A. Pieckiel said: > > I've got an Exabyte M2 connected to my system. I'm running version > > 5.2.1 and have seen this drive do 64K transfers with "systat -vm" > > before (although that was under version 4.9). I can't for the life > > of me find out why I'm only getting 10K per transaction while I'm > > reading from my tape now. At the rate it's going, it will take over > > 10 hours to read through the entire thing. I know the tape drive is > > capable of reading an entire tape in about three hours. > > Tar defaults to a 10k blocksize. When creating tapes, use the 'b' > flag to specify blocksize in 512-byte units. For a 64k blocksize: > "tar cvbf 128 /dev/sa0 /usr" I hit the same 10K limit when I use something like this: # dd if=/dev/nsa0 of=/dev/null bs=64k Is that because tar wrote to the tape with a 10k bock size that even dd can't read larger blocks? That doesn't seem likely to me, but then again, what do I know?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041007105050.GA7349>