Harry Putnam <[email protected]> writes: D> Afshin Salek <[email protected]> writes: > >> If you have any sharesmb property set on you ZFS file systems >> turn them off, if you have any CIFS shares defined with sharemgr >> remove them then reboot the system and try again. > > All I find of smb looks innocuous and not sure how to turn them off > anyway. Its already off on the root of the fs > > zfs get all rpool/z1|grep smb > rpool/z1 sharesmb off default > > But further down the name is all that's set. > > zfs get all rpool/z1/hosts|grep smb > rpool/z1/hosts sharesmb name=Zhosts local > > zfs get all rpool/z1/hosts/mob1|grep smb > rpool/z1/hosts/mob1 sharesmb name=Zmob1 local
Haaa, just an update I finally got the darn thing removed. It required zfs destroy from bottom dir on up. Not sure why destroy -r wasn't doing the job but probably more of my bungling. Any way I have a clean slate as of now. One thing concerns me and its kind of involved. I changed the UID on user reader trying to make the uid match the same users UID on a linux nfs server. I had the notion that users UID had to match for osol user reader to be able to mount an nfs share served from a linux server and be able to write there. It didn't make one whit of difference having matching UIDs, or at least I saw no change in being able to mount the nfs share. I'd be happy enough about the nfs stuff to have root do the mounting long as user reader can write there. I haven't been able to accomplish that yet, but wondering now if changing readers uid is more trouble than its worth. Where I think its showing up is in /var/smb/smbpasswd which still shows reader uid as 101, even though `id reader' shows 1000. I've reset readers passwd repeatedly but the UID in /var/smb/smbpasswd never shows any change. /etc/passwd shows the new UID and there have been multiple reboots that should have caught up what ever is missing there. /var/smb/smbpasswd reader:101::7FC008F63199AC24F0510A57B4EA4560 /etc/passed reader:x:1000:10:Harry Putnam:/export/home/reader:/bin/bash I'm not really sure that is a problem but it seem pretty likely. _______________________________________________ cifs-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
