mandrakeexpert incident 32252 forwarded to cooker. sorry no emailaddress is listed in the mandrakeexpert database for the originator of this report.
----quoted text below---- jbuberel : 18/09 11:21 : Incident created I have a client machine running 9.0rc1. It mounts it's home directory from an NFS server on the local network. This appears to be working just fine. I can log into a KDE session on the client machine without any problems, and my home directory is mounted correctly. However, when I attempt to log into a GNOME session on the client machine, the login sequence stops and reports: ...could not lock the file '/home/jason/.gconf-test-locking-file'... ...a common cause is that the nfs.lockd service has been disabled... Here is the /etc/fstab on the client: server:/home /home nfsr w,lock,hard,rsize=8192,wsize=8192,timeo=14,intr 0 0 Here is the /etc/exports entry on the server: /home 192.168.1.2(rw) 192.168.1.3(rw) Here is the rpcinfo -p output on the client: program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100021 1 udp 32768 nlockmgr 100021 3 udp 32768 nlockmgr 100021 4 udp 32768 nlockmgr 391002 2 tcp 32768 sgi_fam And here is the rpcinfo -p output on the server: program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 391002 2 tcp 32769 sgi_fam 100011 1 udp 1011 rquotad 100011 2 udp 1011 rquotad 100011 1 tcp 1014 rquotad 100011 2 tcp 1014 rquotad 100005 1 udp 1029 mountd 100005 1 tcp 32770 mountd 100005 2 udp 1029 mountd 100005 2 tcp 32770 mountd 100005 3 udp 1029 mountd 100005 3 tcp 32770 mountd 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100021 1 udp 1030 nlockmgr 100021 3 udp 1030 nlockmgr 100021 4 udp 1030 nlockmgr 100024 1 udp 30617 status 100024 1 tcp 3359 status I have checked the firewall logs, and nothing (TCP or UDP) is bieing blocked between the two machines. And as I stated above, the NFS mounting seems to be working just fine for non-Gnome session logins. Any help would be appreciated. Thanks, jason jbuberel : 18/09 11:51 : More info provided This looks to be related to a failure in rpc.statd. The /etc/init.d/nfslock script is configured to run at startup time. This script should launch both rcp.lockd and rpc.statd. However, rpc.statd will not start successfully. When starting rpc.statd manually from the command line, the following errors are reported: [root@jasmine root]# /sbin/rpc.statd -F -d 09/18/2002 23:41:07 rpc.statd[0]: Flags: No-Daemon Log-STDERR 09/18/2002 23:41:07 rpc.statd[5199]: Version 1.0.1 Starting 09/18/2002 23:41:07 rpc.statd[5199]: Cannot access current directory after dropping privs: access() The reason for this problems is that the default ownership of files in the /var/lib/nfs/statd directory is 'root', but statd is configured to run as 'rpcuser'. Could someone file this as a bugreport? It would be good to get this fixed in 9.0rc2. or the final 9.0 release. -jason -----end quoted text----- -- Alan