Hi, It seems to me that this isn't a problem of the NFS lockd directly but of the NFS statd which is patched into the kernel (kstatd) in Suse Linux (other distros use a separate rpc.statd process). Try to boot your server with the original Suse-Kernel and check if it works. Usually the kstatd-patch can be found in the kernel-source.src.rpm of recent Suse kernels. Hope this helps... Greetings and merry christmas Jens Siebert Jörg Spilker wrote:

Hello,

i´ve some NFS filelocking problem between my SuSE 10.0 based clients and the 9.3 based server. /home is located on the server (with SuSE 9.3 and a recent kernel (2.6.15)). The clients are equiped with SuSE 10.0 and the default 2.6.13 kernel.

Applications which are using NFS filelocking on /home (examples: eclipse, skype, banshee) are not working.

The server log shows the following message:

Dec 19 21:35:55 daolin kernel: lockd: cannot monitor 192.168.9.10

I know now (after googling for some minutes), that the reason for the message is, that the lockd on the server cannot contact the status daemon on the client site.

rpcinfo on Server/Client shows the following info:

Server: Suse Linux 9.3 with Kernel 2.6.15 daolin:/var/log # rpcinfo -p daolin Program Vers Proto Port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100004 2 udp 795 ypserv 100004 1 udp 795 ypserv 100004 2 tcp 798 ypserv 100004 1 tcp 798 ypserv 100007 2 udp 814 ypbind 100007 1 udp 814 ypbind 100007 2 tcp 817 ypbind 100007 1 tcp 817 ypbind 100009 1 udp 834 yppasswdd 600100069 1 udp 849 fypxfrd 600100069 1 tcp 851 fypxfrd 100011 1 udp 858 rquotad 100011 2 udp 858 rquotad 100011 1 tcp 877 rquotad 100011 2 tcp 877 rquotad 391002 2 tcp 601 sgi_fam 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 4 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100003 4 tcp 2049 nfs 100021 1 udp 2262 nlockmgr 100021 3 udp 2262 nlockmgr 100021 4 udp 2262 nlockmgr 100021 1 tcp 3171 nlockmgr 100021 3 tcp 3171 nlockmgr 100021 4 tcp 3171 nlockmgr 100005 1 udp 724 mountd 100005 1 tcp 739 mountd 100005 2 udp 724 mountd 100005 2 tcp 739 mountd 100005 3 udp 724 mountd 100005 3 tcp 739 mountd

Client: Suse Linux 10.0 with Kernel 2.6.13-15-default

daolin:/var/log # rpcinfo -p lotus Program Vers Proto Port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 391002 2 tcp 749 sgi_fam 100007 2 udp 810 ypbind 100007 1 udp 810 ypbind 100007 2 tcp 811 ypbind 100007 1 tcp 811 ypbind 100024 1 udp 32779 status 100021 1 udp 32779 nlockmgr 100021 3 udp 32779 nlockmgr 100021 4 udp 32779 nlockmgr 100024 1 tcp 32803 status 100021 1 tcp 32803 nlockmgr 100021 3 tcp 32803 nlockmgr 100021 4 tcp 32803 nlockmgr

i´m not sure but i think that status is the service which the server can´t contact. There are no firewall rules active during my tests.

I can telnet the tcp port for status from daolin.

What looks strange for me are the high port numbers for lockmgr and status on the client.

I really don´t know what is wrong.

Greetings, Jörg

--------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-unsubscribe@opensuse.org For additional commands, e-mail: opensuse-help@opensuse.org