In fact I run the rcnfsserver script. It lists this: # rcnfsserver status Checking for kernel based NFS server: idmapd running mountd unused statd unused nfsd unused nfsserver.service - LSB: Start the kernel based NFS daemon Loaded: loaded (/etc/init.d/nfsserver) Active: failed (Result: exit-code) since Thu 2014-10-30 13:57:56 UTC; 1h 23min ago Process: 2983 ExecStart=/etc/init.d/nfsserver start (code=exited, status=7) I ran the systemd journal command to see if I could get more info. And it was where the mountd error info was provided. I have the following installed: nfs-kernel-server-1.2.8-4.13.1.i586 rpcbind-0.2.0_git201103171419-12.1.2.i586 I looked at your bugzilla reference but it was not clear what I could do about it. When I ran the script with tracing, it did not tell too much. It listed the values of various variables, and then said: # Check if the service is used under systemd but not started with if test -z "$SYSTEMD_NO_WRAP" && /usr/bin/mountpoint -q /sys/fs/cgroup/systemd; then if test $PPID -ne 1 -a $# -eq 1 ; then _rc_base= _sd_opts= case "$0" in /etc/init.d/boot.*) _rc_base=${0##*/boot.} ;; /etc/init.d/*|/etc/rc.d/*) _rc_base=${0##*/} ;; */rc*) if test -L "$0"; then _rc_base=`readlink "$0"` _rc_base=${_rc_base##*/} case "$_rc_base" in boot.*) _rc_base=${_rc_base#boot.} esac else _rc_base=${0##*/rc} fi ;; esac case "$1" in status) SYSTEMD_NO_WRAP=1 "$0" "$1" _sd_opts='--lines=0 --full --output=cat' ;; start|stop|reload|restart|try-restart|force-reload) echo "redirecting to systemctl $1 ${_rc_base}.service" 1>&2 ;; *) unset _rc_base esac if test -n "$_rc_base" -a -x /usr/bin/systemctl ; then exec /usr/bin/systemctl $_sd_opts $1 "${_rc_base}.service" fi unset _rc_base _sd_opts fi if test -z "$REDIRECT" -a -x /sbin/showconsole ; then REDIRECT="$(/sbin/showconsole 2>/dev/null)" test -z "$CONSOLE" && CONSOLE=/dev/console export REDIRECT CONSOLE fi fi ++ test -z '' ++ /usr/bin/mountpoint -q /sys/fs/cgroup/systemd ++ test 2818 -ne 1 -a 1 -eq 1 ++ _rc_base= ++ _sd_opts= ++ case "$0" in ++ _rc_base=nfsserver ++ case "$1" in ++ echo 'redirecting to systemctl start nfsserver.service' redirecting to systemctl start nfsserver.service ++ test -n nfsserver -a -x /usr/bin/systemctl ++ exec /usr/bin/systemctl start nfsserver.service Job for nfsserver.service failed. See 'systemctl status nfsserver.service' and 'journalctl -xn' for details. No clues that I recognize. Roger Oberholtzer Ramböll RST Office: +46 (0)10-615 6020 Mobile: +46 (0)70-815 1696 roger.oberholtzer@ramboll.se ________________________________________ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden www.rambollrst.se ________________________________________ From: Anton Aylward [opensuse@antonaylward.com] Sent: Thursday, October 30, 2014 2:01 PM To: opensuse@opensuse.org Subject: Re: [opensuse] openSUSE 13.1 mountd failure On 10/30/2014 08:38 AM, Roger Oberholtzer wrote:

I do not see why this is failing. It seems to be failing with mountd. No idea why. Anyone ever seen this?

Roger Oberholtzer

First off, this is not a systemd problem I've encountered it in 11.1 and 11.2 as well :-( Secondly, the kernel NFS server is an old sysvinit script. # rpm -ql nfs-kernel-server /etc/init.d/nfsserver /etc/sysconfig/SuSEfirewall2.d/services/nfs-kernel-server /usr/sbin/exportfs /usr/sbin/nfsdcltrack /usr/sbin/rcnfsserver /usr/sbin/rpc.mountd /usr/sbin/rpc.nfsd /usr/sbin/rpc.svcgssd /usr/share/man/man5/exports.5.gz /usr/share/man/man7/nfsd.7.gz /usr/share/man/man8/exportfs.8.gz /usr/share/man/man8/mountd.8.gz /usr/share/man/man8/nfsd.8.gz /usr/share/man/man8/nfsdcltrack.8.gz /usr/share/man/man8/rpc.mountd.8.gz /usr/share/man/man8/rpc.nfsd.8.gz /usr/share/man/man8/rpc.svcgssd.8.gz /usr/share/omc/svcinfo.d/nfs-kernel-server.xml /var/lib/nfs/etab /var/lib/nfs/rmtab /var/lib/nfs/xtab See, /etc/init.d/nfsserver and no systemd unit. What is happening in your trace report is that systemd is running in backwards compatibility mode to make use of the sysvinit script. I'd start the debugging by running the init script in debug mode sh -xv /etc/init.d/nfsserver start My personal suspicions, based on what I went though with 11.x, is that the RPC system isn't there. Check for the portmapper. The nice thing about when stuff is fully converted to systemd is that you can make the dependencies explicit. While its not impossible to do that with scripts it makes them a bit unwieldy. However ... It may also be a nfs v4 problem https://bugzilla.redhat.com/show_bug.cgi?id=443625 -- Never hire or promote in your own image. It is foolish to replicate your strength and idiotic to replicate your weakness. It is essential to employ, trust, and reward those whose perspective, ability, and judgment are radically different from yours. It is also rare, for it requires uncommon humility, tolerance, and wisdom. - Dee W. Hock, Fast Company -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Not sure if this is what you meant: linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # ls -l total 0 -rw-r--r-- 1 root root 0 Oct 30 10:58 cgroup.clone_children --w--w--w- 1 root root 0 Oct 30 10:58 cgroup.event_control -rw-r--r-- 1 root root 0 Oct 30 16:06 cgroup.procs -rw-r--r-- 1 root root 0 Oct 30 10:58 notify_on_release -rw-r--r-- 1 root root 0 Oct 30 10:58 tasks linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat tasks 2597 linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat notify_on_release 1 linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat cgroup.procs 2597 linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat cgroup.event_control cat: cgroup.event_control: Invalid argument linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat cgroup.clone_children 0 I suspect you are interested in something else. Roger Oberholtzer Ramböll RST Office: +46 (0)10-615 6020 Mobile: +46 (0)70-815 1696 roger.oberholtzer@ramboll.se ________________________________________ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden www.rambollrst.se ________________________________________ From: Anton Aylward [opensuse@antonaylward.com] Sent: Thursday, October 30, 2014 3:10 PM To: opensuse@opensuse.org Subject: Re: [opensuse] openSUSE 13.1 mountd failure No need to cc me on your replies; I subscribe to this list. On 10/30/2014 09:54 AM, Roger Oberholtzer wrote:

++ echo 'redirecting to systemctl start nfsserver.service' redirecting to systemctl start nfsserver.service ++ test -n nfsserver -a -x /usr/bin/systemctl ++ exec /usr/bin/systemctl start nfsserver.service Job for nfsserver.service failed. See 'systemctl status nfsserver.service' and 'journalctl -xn' for details.

And what is in your unit nfsserver.service ? -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

I disabled IPv6 and rebooted. Still no joy. Must I disable IPv6 for rpcbind as well? I cannot see why this is failing. Roger Oberholtzer ________________________________________ From: Roger Oberholtzer Sent: Thursday, October 30, 2014 3:56 PM To: opensuse@opensuse.org Subject: RE: [opensuse] openSUSE 13.1 mountd failure If I check how far the nfsserver script gets, it seems to be to where it starts /usr/sbin/rpc.mountd I see that it runs the command with no options. So the possible parameters to the script are not defined. So I tried running it in the foreground as: /usr/sbin/rpc.mountd -F And I get:: rpc.mountd: mountd: could not create listeners Which is the error. So it seems the the items previous to this in the script have been started. But someone is not happy. I wonder if IPv6 is involved. I do not use it, and it is however openSUSE sets it up. If not using YaST, where can I disable this? Roger Oberholtzer ________________________________________ From: Roger Oberholtzer Sent: Thursday, October 30, 2014 3:18 PM To: opensuse@opensuse.org Subject: RE: [opensuse] openSUSE 13.1 mountd failure Not sure if this is what you meant: linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # ls -l total 0 -rw-r--r-- 1 root root 0 Oct 30 10:58 cgroup.clone_children --w--w--w- 1 root root 0 Oct 30 10:58 cgroup.event_control -rw-r--r-- 1 root root 0 Oct 30 16:06 cgroup.procs -rw-r--r-- 1 root root 0 Oct 30 10:58 notify_on_release -rw-r--r-- 1 root root 0 Oct 30 10:58 tasks linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat tasks 2597 linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat notify_on_release 1 linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat cgroup.procs 2597 linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat cgroup.event_control cat: cgroup.event_control: Invalid argument linux:/sys/fs/cgroup/systemd/system.slice/nfsserver.service # cat cgroup.clone_children 0 I suspect you are interested in something else. Roger Oberholtzer ________________________________________ From: Anton Aylward [opensuse@antonaylward.com] Sent: Thursday, October 30, 2014 3:10 PM To: opensuse@opensuse.org Subject: Re: [opensuse] openSUSE 13.1 mountd failure No need to cc me on your replies; I subscribe to this list. On 10/30/2014 09:54 AM, Roger Oberholtzer wrote:

++ echo 'redirecting to systemctl start nfsserver.service' redirecting to systemctl start nfsserver.service ++ test -n nfsserver -a -x /usr/bin/systemctl ++ exec /usr/bin/systemctl start nfsserver.service Job for nfsserver.service failed. See 'systemctl status nfsserver.service' and 'journalctl -xn' for details.

And what is in your unit nfsserver.service ? -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Thanks Per! I saw that disabling IPv6 did not make a difference. I did it only because I saw complaints when straceing rpc.mountd, whihc seems the one that will not start.

From your list, I have only this running:

/usr/sbin/rpc.idmapd -p /var/lib/nfs/rpc_pipefs I did not expect to see rpc.mountd as I see that is failing. I wonder why rpcbind is not running. I do not see it getting started in the nfsserver script. Maybe a package is missing? Roger Oberholtzer Ramböll RST Office: +46 (0)10-615 6020 Mobile: +46 (0)70-815 1696 roger.oberholtzer@ramboll.se ________________________________________ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden www.rambollrst.se ________________________________________ From: Per Jessen [per@computer.org] Sent: Thursday, October 30, 2014 4:41 PM To: opensuse@opensuse.org Subject: RE: [opensuse] openSUSE 13.1 mountd failure Roger Oberholtzer wrote:

I disabled IPv6 and rebooted. Still no joy. Must I disable IPv6 for rpcbind as well? I cannot see why this is failing.

Hi Roger I have a brandnew openSUSE factory system with an NFS-server, it's working fine. No need to fiddle with ipv6. I'll be happy to compare notes. The following daemons are running: /usr/sbin/rpc.idmapd -p /var/lib/nfs/rpc_pipefs /sbin/rpcbind -w -f /usr/sbin/rpc.mountd /usr/sbin/rpc.statd --no-notify /etc/exports: /home/office12 office12.local.net(rw,sync,no_subtree_check,no_root_squash) -- Per Jessen, Zürich (12.7°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

В Thu, 30 Oct 2014 16:29:25 +0000 Roger Oberholtzer пишет:

Hmm. I see that rpcbind is installed. But it is not started. If I do so by hand (run the rpcbind command), NFS starts!

I see that rpcbind is managed by systemd, while NFS starts via a SysV script. A dependency seems broken.

Yes, rpcbind is disabled by default. You need to enable it: systemctl enable rpcbind -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Yes. I read a similar description that listed the command line method (my first choice) as well as the Yast method. In this system, I tried the Yast method as that has always worked. The Yast method does not seem to work. It seems it does not enable/start rpcbind. Or if it does, it fails. I will be putting together the systems today and I can hopefully report that they are working - as long as one enabled rpcbind by hand. Roger Oberholtzer Ramböll RST Office: +46 (0)10-615 6020 Mobile: +46 (0)70-815 1696 roger.oberholtzer@ramboll.se ________________________________________ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden www.rambollrst.se ________________________________________ From: Anton Aylward [opensuse@antonaylward.com] Sent: Thursday, October 30, 2014 10:53 PM To: opensuse@opensuse.org Subject: Re: [opensuse] openSUSE 13.1 mountd failure On 10/30/2014 01:14 PM, Andrei Borzenkov wrote:

В Thu, 30 Oct 2014 16:29:25 +0000 Roger Oberholtzer пишет:

...

Hmm. I see that rpcbind is installed. But it is not started. If I do so by hand (run the rpcbind command), NFS starts!

I see that rpcbind is managed by systemd, while NFS starts via a SysV script. A dependency seems broken.

Yes, rpcbind is disabled by default. You need to enable it:

systemctl enable rpcbind

As in http://www.unixmen.com/setup-nfs-server-opensuse-13-1/ -- Genius is one percent inspiration and ninety-nine percent <strike>perspiration<>/strike> using google to see arch for relevant articles. - Thomas Alva Edison -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org N�����r��y隊Z)z{.�ﮞ˛���m�)z{.��+�:�{Zr�az�'z��j)h���Ǿ� ޮ�^�ˬz��

Good. ISTR having to start rpcbind manually, but I'm not sure. It's certainly running, but not enabled.

Is rpcbind.socket enabled? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Fri, Oct 31, 2014 at 11:22 AM, Per Jessen wrote:

Andrei Borzenkov wrote:

...

...

Good. ISTR having to start rpcbind manually, but I'm not sure. It's certainly running, but not enabled.

Well, your output shows that it is enabled.

Loaded: loaded (/usr/lib/systemd/system/rpcbind.service; enabled) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 10/30/2014 10:18 AM, Roger Oberholtzer wrote:

I suspect you are interested in something else.

Yes. Read systemd.unit (5) and systemd.service (5) -- It is not a question of how well each process works, the question is how well they all work together. - Lloyd Dobens and Clare Crawford-Mason Thinking About Quality -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org