hmm, on Mon, Aug 04, 2014 at 01:15:03PM +0000, Stuart Henderson said that > On 2014-08-03, Carsten Kunze <[email protected]> wrote: > > I use sshfs to synchronize a filesystem of 15 GB between two machines. > > Read access seems to be ok but on writing the mount point > > does not seem to work anymore. Error message of cp(1) is > > > > No such file or directory > > > > ls(1) to the mount point gives the same message. > > (Nothing in /var/log/messages) > > > > Is it a known problem for sshfs that the mount point may > > disappear? (OpenBSD 5.5 amd64) > > > > --Carsten > > > > > > There were some problems with FUSE that were fixed post-5.5, perhaps > you ran into one of those. (There are still certainly some missing > features and possibly also other problems in the existing code, but > no recent reports). > > If you still see such problems on -current/forthcoming 5.6 then the > best option would be to make a proper problem report, with log > information / output from a run with FUSE_DEBUG set, backtraces > from the sshfs process if it crashes, etc.
actually i am seeing fuse regression with ntfs-3g in -current. i was able to copy ~1TB of data from external usb drives formatted ntfs in 5.5. in -current the process accessing the drive will inevitably spin and is not possible to kill. i was looking for a way to make a debug build and make a more detailed report. of course it could be also ntfs-3g but it wasn't updated since april. -f -- if money doesn't grow on trees, why is it green?
