I am fishing a bit here, but I have no_root_squash as an option in my exports (server) file. When the client tries to mount the directory automatically at boot time through the entry in /etc/fstab I believe the request is identified as coming from root. See the "User ID Mapping" section in man (5) exports about this. On 10/23/2011 08:36 PM, Gerhard Magnus wrote: > I can ping (and ssh) the server from the client. > When I restart nfs and nfslock on the server I get this in the > server /var/log/messages: > PuteF kernel: [25572.902466] NFSD: Using /var/lib/nfs/v4recovery as the > NFSv4 state recovery directory > PuteF kernel: [25572.902504] NFSD: starting 90-second grace period > PuteF rpc.mountd[8699]: Version 1.2.4 starting > But on the client, mount /home/magnusg/Music returns: > mount.nfs: access denied by server while mounting > 192.168.1.14:/home/magnusg/Music > > a couple more things > (1) on the client, showmount -e 192.168.1.14 (the server) > Export list for 192.168.1.14: (shows nothing) > But on the server, my /etc/exports file contains: > /home/magnusg/Music 192.168.1.12,192.168.1.13(rw,insecure,sync,nohide) > (2) also on the client, trying to mount with mount.nfs: > mount.nfs PuteF:/home/magnusg/Music /home/magnusg/Music -v > mount.nfs: timeout set for Sun Oct 23 17:25:32 2011 > mount.nfs: trying text-based options > 'vers=4,addr=192.168.1.14,clientaddr=192.168.1.13' > mount.nfs: mount(2): Permission denied > mount.nfs: access denied by server while mounting > PuteF:/home/magnusg/Music > > OK, my only contact with people who use Linux these days is on this > list, so I have no idea what the current "lore" is. Does anybody > actually use NFS anymore? If not, is there any alternative to NFS > besides using ssh? > > ------------------------------- > > Restating the problem from yesterday: > > On the server, running FC15 (192.168.1.14): > (1) My /etc/exports file looks like this: > /home/magnusg/Music 192.168.1.12,192.168.1.13(rw,insecure,sync,nohide) > (2) Using the system-config-nfs "General Options tab" I have "Allow > connections from port 1024 and higher" checked. > (3) I have services nfs and nfslock running on levels 3,4,5. > (4) In /etc/sysconfig/nfs I've set these ports: > RQUOTAD_PORT=4000 > LOCKD_TCPPORT=4001 > LOCKD_UDPPORT=4001 > MOUNTD_PORT=4002 > STATD_PORT=4003 > (5) In the firewall I have these ports open: > NFSV4 2049 (tcp) > 4000-4003 (tcp and udp) > 111 (tcp and udp) > > on the client, running FC13 (192.168.1.13): > (1) I added this to /etc/fstab: > 192.168.1.14:/home/magnusg/Music /home/magnusg/Music nfs > rw,auto,hard,intr,bg 0 0 > (2) Services netfs, nfslock and rpcbind are running on levels 3,4,5. > > When I boot the client and get to the "Mounting NFS filesystems" section > I don't get error messages -- but I do see this: > mount.nfs: backgrounding 192.168.1.14:/home/magnusg/Music > : mount options > "hard,intr,bg,vers=4,addr=192.168.1.14,clientaddr=192.168.1.13 > /home/magnusg/Music remains unmounted > > -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
