Ludovic Courtès <[EMAIL PROTECTED]> writes: > Hi, > > On Mon, May 05, 2003 at 12:09:10PM +0200, Johannes Rohr wrote: > > since my box has network access now, I was very curious to try out the > > Hurd's famous translators. But both the nfs and the ftpfs translator > > do not seem to work. Do I assume correctly that for nfs, all I have to > > do is > > > > settrans /some/place /hurd/nfs host:/someexport ? > > The above line changes the passive translator settings for the > "/some/place" node. In other words, it does not actually start the > translator until you access the node. > > For debugging purposes, you should use "settrans -a" which makes the > translator active; adding "-P" will make settrans wait until you press a > key. This way, you have enough time to get the PID of the translator, > run "gdb /hurd/nfs <PID>", and press a key in order to start the > translator.
Been there, done that. Still, '-P' doesn't seem to work for me as advertised. The translator dies immediately. What really surprises me is that the ftp translator dies _momentarily_, as if it even wouldn't bother to contact the server. At the same time I've seen that I can ftp with it to a neighbouring box (while any attempts to set up an ftp translator for any server out in the Internet fail with 100 % reproducibility. I wonder if there is any way to intercept the communication between the ftp translator and the server... Thanks, Johannes -- ~/.signature under construction