User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.8.1.3) Gecko/20070413 Firefox/2.0.0.3 Build Identifier: Hi, My /home partition is an nfs export on a server (clk01) and my computer (clk02) mounts it at boot so all my users have centralized "homes". Everything works as expected there. I have enabled rpc.lockd and rpc.statd in my /etc/rc.conf because eclipse and openoffice.org-2.2.0 require "true" file locking. When enable, the said applications also work as expected. Having outgrown KDE and gnome and installed Xfce-4.4.1 as my DE, I was happy until I found out that when rpc.lockd is enabled, Thunar (the file manager) will freeze as rpc.lockd is sent into a seemingly infinite loop. Though a kill on rpc.lockd will work, it will break anything that requires locking and turn Thunar into an unkillable (STOP) state, requiring a reboot. Here's what I get in my debug.log when both deamons are started with "-d 2" flags: Apr 15 11:08:21 clk02 rpc.statd: unmon_all for host: NFS NLM prog: 0 ver: 0 proc: 0 Apr 15 11:10:40 clk02 rpc.lockd: lock request: V4: write to 192.168.1.10 Apr 15 11:10:40 clk02 rpc.lockd: nlm4_lock_res from clk01.clkroot.net Apr 15 11:10:40 clk02 rpc.lockd: lock answer: pid 883: nlmv4 3 Apr 15 11:11:00 clk02 rpc.lockd: lock request: V4: write to 192.168.1.10 Apr 15 11:11:00 clk02 rpc.lockd: nlm4_lock_res from clk01.clkroot.net Apr 15 11:11:00 clk02 rpc.lockd: lock answer: pid 883: nlmv4 3 Apr 15 11:11:20 clk02 rpc.lockd: lock request: V4: write to 192.168.1.10 Apr 15 11:11:20 clk02 rpc.lockd: nlm4_lock_res from clk01.clkroot.net Apr 15 11:11:20 clk02 rpc.lockd: lock answer: pid 883: nlmv4 3 Apr 15 11:11:40 clk02 rpc.lockd: lock request: V4: write to 192.168.1.10 Apr 15 11:11:40 clk02 rpc.lockd: nlm4_lock_res from clk01.clkroot.net Apr 15 11:11:40 clk02 rpc.lockd: lock answer: pid 883: nlmv4 3 Apr 15 11:12:00 clk02 rpc.lockd: lock request: V4: write to 192.168.1.10 Apr 15 11:12:00 clk02 rpc.lockd: nlm4_lock_res from clk01.clkroot.net Apr 15 11:12:00 clk02 rpc.lockd: lock answer: pid 883: nlmv4 3 And it will go on forever until a reboot or a kill on both deamons. Note that this behavior is only initiated when starting Thunar. Also, remember that Thunar will work correctly when rpc_lockd/rpc_statd is not running. Reproducible: Always Steps to Reproduce: 1. Have /home on a nfs mount 2. Enable rpc_lockd and rpc_statd in /etc/rc.conf 3. Start Thunar
Disregard. The bug isn't on Thunar's side but on rpc_lockd which is buggy. Solution is to mount_nfs with option nolockd and Thunar will work fine.