On Fri, 2002-12-06 at 07:39, Richard Sharpe wrote: > On Thu, 5 Dec 2002, Patrick Welche wrote: > > > The following used to work with several months old code.. > > > > #============================ Share Definitions ============================== > > [prlw1] > > comment = Windoze transfer > > path = /usr/windoze > > valid users = prlw1 > > browseable = no > > writable = yes > > guest ok = no > > > > [homes] > > comment = Home Directories > > browseable = no > > writable = yes > > > > > > Now when I smbclient //localhost/prlw1 I end up in my home directory > > /home/prlw1 rather than in /usr/windoze, so it seems the [prlw1] share > > isn't being matched, and as prlw1 is a valid userid it is then matched > > to [homes]. Any thoughts on how I could debug this? > > Which version of the code? Sounds like someone changed the order of the > test. Previously, it used to check for a share of that name, and if not > found, checked for [homes] and then [printers].
Yep, I think I changed the order of that test - because we use use the 'snum' attached to the vuid if possible, so as to avoid expensive lookups. The correct fix would be to fix lp_add_home() not to overwrite things when it finds that the home dir share already exists. (Because all the damage is actually done there - the order that the search is done doesn't matter any more, as the share is added as session setup time). Unfortunately the fix I was about to commit seems to be the opposite of an earlier 'fix' for a related bug. I'll need to stare at this a bit more... Andrew Bartlett -- Andrew Bartlett [EMAIL PROTECTED] Manager, Authentication Subsystems, Samba Team [EMAIL PROTECTED] Student Network Administrator, Hawker College [EMAIL PROTECTED] http://samba.org http://build.samba.org http://hawkerc.net
signature.asc
Description: This is a digitally signed message part
