According to hdparm -t /dev/hda on both machines I get 20.51 and 54.61 MB/sec, so that shouldn't be the problem. I guess I'll try rebooting them, though that "shouldn't" make any difference. Then I'll try it again
On Friday 05 September 2003 05:41 pm, Martin Larsson wrote: > hmm, i get up to 9,73 MB/sec on a 100base using http from a apache > server > > maybe the HD's cant write faster than ~1mb/sec (altough it > should.....) > > Ernie Schroder wrote: > | OK so I've been burned and am finally doing regular backups. > | There are 6 boxes on the home LAN here and I had a 5 port 10/100 > | hub and a 10 Base 8 port hub so at best 2 machines were stuck at > | a slower connection. This didn't bother me much because there was > | never a great deal of large files being passed. I just used the 8 > | port hub cause the puppy chewed one of my patch cords. This > | morning I went out and got a shiny new 10 port 10/100 switch. So, > | after hooking it up, I did an scp of a stage-3 tarball from 1 > | gentoo box to the other Throughput was 656 kb/sec,virtually the > | same as with the10 Base hub. I restarted net.eth0 on both boxes > | and tried again. This time, the throughput was 971 kb/sec or > | about 1.5 times as fast. Should I expect more? both machines have > | 10/100 nics and the switch shows connection at 100 Base. The > | boxes are both connected to the hub with 6 foot cat5e patch > | cords. Is there some configuration I'm missing do I need to > | reboot? :-( What kind of throughput do others see on a similar > | setup? -- Regards, Ernie 100% Microsoft and Intel free -- [EMAIL PROTECTED] mailing list
