On Mon, 3 Dec 2001 at 9:56am, Stefan G. Weichinger wrote > 1) Not all hosts to be backed up have 100M-NICs inside, so while > dumping them it slows down the connection to 100M-hosts. > > (Getting about 10kB/s right now on the 100M-host, that�s not much ...)
That sounds like a duplex mismatch. Are the 10M boxen on a switch smart enough to deal with this? > 2) I run multiple dumps a day right now because I�m constantly adding > new disks and testing parameters. My tapecycle is 11, my dumpcycle 14, > I have 11 tapes labeled and used them all already, so that they are in > my tapelist. > But amanda is still requesting new tapes. > Is this because I ran multiple dumps per day ? > How I can I tell her to use DAILY01 after DAILY11 explicitly ? What is your runspercycle? Tapecycle *must* be >= (preferebably >) runspercycle. Otherwise, you may be overwriting the *only* full dump of a filesystem you have on tape. If that happens, you're SOL at restore time. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
