What steps will reproduce the problem?
1. On OpenSSH server config set ClientAliveInterval and ClientAliveCountMax to
some values (like 15, 3) and restart the service.
2. Mount a drive via win-sshfs which is connected to the previous OpenSSH
server.
3. Wait acording to the lientAliveInterval and ClientAliveCountMax setting
(like one minute) and the drive becomes unresponsible. Windows is complaining
about missing resources if I try to open the drive.
OpenSSH is writing this to syslog:
sshd[18513]: Timeout, client not responding.
sshd[18511]: pam_unix(sshd:session): session closed for user XXX
What is the expected output? What do you see instead?
The drive should be accessible even if it not used over some time but still
mounted. Therefore win-sshfs should react to the OpenSSH messages caused by the
ClientAliveInterval setting. so the ssh connection would not get closed by the
server.
What version of the product are you using?
win-sshfs: 0.0.1.5
Client OS: Windows 7 Enterprise 32 bit (german)
Server OS: gentoo with latest stable update from 17. Jan. 2013
Ssh server: OpenSSH 5.9_p1-r4
Please provide any additional information below.
By the way here is the short description of that feature from the man page:
ClientAliveInterval
Sets a timeout interval in seconds after which if no data has been received
from the client, sshd(8) will send a message through the encrypted channel to
request a response from the client.
The default is 0, indicating that these messages will not be sent to the client.
This option applies to protocol version 2 only.
Original issue reported on code.google.com by karsten....@gmail.com on 17 Jan 2013 at 10:02
Original issue reported on code.google.com by
karsten....@gmail.comon 17 Jan 2013 at 10:02