Date: Fri, 22 May 1998 09:56:05 -0700 (PDT) From: Jim Van Baalen <vansax@mail.websidestory.com> To: questions@FreeBSD.ORG Subject: rdump question Message-ID: <Pine.BSI.3.96.980522091728.9167A-100000@mail.websidestory.com>
next in thread | raw e-mail | index | archive | help
I have been dumping several machines to local 8mm drives. Dump writes to these drives at about 4mbps. To conserve resources I have been trying to dump to the same model drive on a remote machine using the <hostname>: notation. When I do this dump writes at about .5mbps. Both machines are on dedicated 10bt switched connections and dump is definitely not taxing the 10bt connection. Both machines are 266Mhz which I have seen easily fill a switched 10bt connection. Neither machine is working particulary hard independent of the dump. I have tried replacing the 3COM 3C905 Fast Etherlink XL PCI with the Intel EtherExpress Pro 10/100B because I have had NFS problems while using the 3COM card. It appears the the default block size for dump is 1024 (as opposed to 8192 for NFS) so it I didn't think this would work, but I had the cards... With typical disk capacities of 4-9Gb dumping a drive at this speed can take over a day. Is this expected behavior? Is there a solution to this problem? What is the bottleneck? Jim To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSI.3.96.980522091728.9167A-100000>