Occasionally I have an issue when I try to SCP a file to one of our Cisco devices. When the connection starts, it immediately drops and says "lost connection". I've had this happen when there wasn't enough space on the flash drive but this isn't the case today.
Below is the failure from a 3750X switch and the pertinent config info from the switch.
laptop:C3750X user$ scp c3750e-universalk9-tar.150-2.SE4.tar cisco@10.2.55.149:c3750e-universalk9-tar.150-2.SE4.tar
Password:
c3750e-universalk9-tar.150-2.SE4.tar 0% 0 0.0KB/s --:-- ETAConnection to 10.2.55.149 closed by remote host.
lost connection
laptop:C3750X user$ ssh cisco@10.2.55.149
Password:
temp#
temp#sh run | sec aaa
aaa new-model
aaa authentication login default local
aaa authorization exec default local
aaa session-id common
temp#sh run | sec ssh|scp
ip ssh version 2
ip scp server enable
transport input ssh
temp#sh run | sec line
line con 0
line vty 5 15
transport input ssh
temp#dir
Directory of flash:/
526 -rwx 616 Mar 30 2011 01:39:31 +00:00 vlan.dat
527 -rwx 2072 Mar 30 2011 01:40:52 +00:00 multiple-fs
528 drwx 512 Mar 30 2011 03:23:11 +00:00 update
529 -rwx 2955 Mar 30 2011 01:40:52 +00:00 config.text
530 -rwx 3561 Mar 30 2011 01:40:52 +00:00 private-config.text
2 drwx 512 Mar 1 1993 00:12:47 +00:00 c3750e-universalk9npe-mz.150-2.SE3
57671680 bytes total (31971328 bytes free)
temp#
I forgot to include the user section but cisco/cisco is a temp account with privilege 15.
Any ideas why SCP is failing when SSH works just fine?
EDIT: SCP debug output
temp#
Mar 30 03:33:37.452: SCP: [22 -> 10.0.13.120:60706] send <OK>
Mar 30 03:33:37.687: SCP: [22 <- 10.0.13.120:60706] recv C0644 25548800 c3750e-universalk9-tar.150-2.SE4.tar
Mar 30 03:33:37.704: SCP: [22 -> 10.0.13.120:60706] send <OK>
Mar 30 03:33:38.400: SCP: [22 -> 10.0.13.120:60706] send Write failed
temp#
-v
on the SCP command to get a more verbose output of what's happening.