hello,

i've run into the following problem:

i have three hosts debian_a,debian_b and mac_c. debian_a is a server running ssh daemon, debian_b and mac_c want to scp to it. debian_a and debian_b are running linux, debian testing. mac_c is running mac os x 10.2.8. the hosts are all in the same subnet. (which is 'virtual' - that's what i was told, i'm not sure what it means. fact is, that the hosts are in quite different parts of the building, server room and offices, but (virtually?) grouped together in subnets.)

scp works the way i want, but the transfer rate from mac_c to debian_a is ridiculous (ca. 20 KB/s and frequent stalls, as opposed to the transferrate of ca. 1 MB from debian_b to debian_a). the strange thing is, that the transfer rate from mac_c to debian_b is quite the same as from debian_b to debian_a. i therefore think that it's got something to do with the network, there must be some obstacle between mac_c and debian_a.

since i'm not at all savvy about these network things, i'd really appreciate if someone has pointers for me as to how to troubleshoot this situation, i'm not quite sure where to start.

i've tried with scp -v from both debian_b and mac_c, but i didn't find any relevant differences.

mac_c to debian_a scp -v output:


Password:
debug1: ssh-userauth2 successful: method keyboard-interactive
debug1: fd 4 setting O_NONBLOCK
debug1: fd 5 setting O_NONBLOCK
debug1: channel 0: new [client-session]
debug1: send channel open 0
debug1: Entering interactive session.
debug1: ssh_session2_setup: id 0
debug1: Sending command: scp -v -r -p -d -t ~
debug1: channel request 0: exec
debug1: channel 0: open confirm rwindow 0 rmax 32768
Sending file modes: C0644 13411140 file.tif
file.tif 1% | | 200 KB 03:13 ETA
<i abort with strg-c>
debug1: channel_free: channel 0: client-session, nchannels 1
debug1: fd 0 clearing O_NONBLOCK
debug1: fd 1 clearing O_NONBLOCK
Killed by signal 2.
debug1: Calling cleanup 0xdd04(0x0)
debug1: Calling cleanup 0x17d98(0x0)

debug1: compress outgoing: raw data 197185, compressed 152111, factor 0.77
debug1: compress incoming: raw data 254, compressed 135, factor 0.53

traceroute to debian_a shows that the connection is direct:

mac_c$ traceroute debian_a
1 debian_a (123.456.789.101) 0.561 ms 0.251 ms 0.218 ms

thanks in advance, please tell me if you need any more info.

regards,

rdorte

How is your network set up? Put your architecture here as well as any devices along the way.

thanks for the quick answer.

as i pointed out, the hosts are all in the same subnet and have direct access to each other, no gateways in between. but the subnet is virtual, i.e. they actually are behind different routers and the like, but this is (or should be) invisible to the hosts. i'm afraid i don't know much about this, since it's not administered by me.

the only thing i know is that mac_c is behind a switch, client debian_b is not. but another machine behind the same switch gets good transfer rates to the server debian_a.

do you suppose the problem is related mainly to the network?

regards and thanks,

rdorte