Pogoplug E02 : I can't start NFS daemon at boot with systemd

[ueghio]

Hi guys,yesterday arrived my pogoplug v2 and I immediately installed Arch linux. I'm a newbie arch user but everything works fine. Everything except for nfs : the service doesn't automatically start at boot.
I think this could be a problem with ntpd syncing because It returns this error message:
$this->bbcode_second_pass_code('', '[marco@pogoplug ~]$ sudo systemctl status nfsd
[sudo] password for marco:
nfsd.service - NFS Server Daemon
Loaded: loaded (/usr/lib/systemd/system/nfsd.service; disabled)
Active: failed (Result: exit-code) since Thu 1970-01-01 01:00:27 CET; 44 years 1 months ago
Docs: man:rpc.nfsd(8)
Process: 165 ExecStopPost=/usr/bin/exportfs -a -u (code=exited, status=0/SUCCESS)
Process: 161 ExecStop=/usr/bin/rpc.nfsd 0 (code=exited, status=0/SUCCESS)
Process: 157 ExecStartPost=/usr/bin/exportfs -a (code=exited, status=1/FAILURE)
Process: 130 ExecStart=/usr/bin/rpc.nfsd $NFSD_OPTS $NFSD_COUNT (code=exited, status=0/SUCCESS)
Main PID: 130 (code=exited, status=0/SUCCESS)
CGroup: /system.slice/nfsd.service

Jan 01 01:00:25 pogoplug systemd[1]: Starting NFS Server Daemon...
Jan 01 01:00:26 pogoplug rpc.nfsd[130]: rpc.nfsd: Setting version failed: errno 22 (Invalid argument)
Jan 01 01:00:27 pogoplug exportfs[157]: exportfs: /srv/nfs4/hdd does not support NFS export
Jan 01 01:00:27 pogoplug exportfs[157]: exportfs: /srv/nfs4 does not support NFS export
Jan 01 01:00:27 pogoplug systemd[1]: nfsd.service: control process exited, code=exited status=1
Jan 01 01:00:27 pogoplug systemd[1]: Failed to start NFS Server Daemon.
Jan 01 01:00:27 pogoplug systemd[1]: Unit nfsd.service entered failed state.')
Anyway after the boot date and time are set correctly.
I have to start manually nfs after the boot with the following command
$this->bbcode_second_pass_code('', 'sudo systemctl start rpc-idmapd.service rpc-mountd.service')
and the problem disappear
$this->bbcode_second_pass_code('', 'marco@pogoplug ~]$ sudo systemctl status rpc-idmapd.service rpc-mountd.service
rpc-idmapd.service - NFS v4 ID-Name Mapping Daemon
Loaded: loaded (/usr/lib/systemd/system/rpc-idmapd.service; enabled)
Active: active (running) since Thu 1970-01-01 01:00:27 CET; 44 years 1 months ago
Docs: man:rpc.idmapd(8)
man:idmapd.conf(5)
Process: 168 ExecStart=/usr/bin/rpc.idmapd $IDMAPD_OPTS (code=exited, status=0/SUCCESS)
Main PID: 170 (rpc.idmapd)
CGroup: /system.slice/rpc-idmapd.service
└─170 /usr/bin/rpc.idmapd

Jan 01 01:00:27 pogoplug systemd[1]: Started NFS v4 ID-Name Mapping Daemon.
Feb 13 22:25:28 pogoplug systemd[1]: Started NFS v4 ID-Name Mapping Daemon.

rpc-mountd.service - NFS Mount Server Daemon
Loaded: loaded (/usr/lib/systemd/system/rpc-mountd.service; enabled)
Active: active (running) since Thu 2014-02-13 22:25:29 CET; 41s ago
Docs: man:rpc.mountd(8)
Process: 1018 ExecStart=/usr/bin/rpc.mountd $MOUNTD_OPTS (code=exited, status=0/SUCCESS)
Main PID: 1019 (rpc.mountd)
CGroup: /system.slice/rpc-mountd.service
└─1019 /usr/bin/rpc.mountd

Feb 13 22:25:29 pogoplug systemd[1]: Starting NFS Mount Server Daemon...
Feb 13 22:25:29 pogoplug systemd[1]: Started NFS Mount Server Daemon.
Feb 13 22:25:29 pogoplug rpc.mountd[1019]: Version 1.2.9 starting')
Have you got any idea about how to solve this?
Thanks in advance. Marco

[Geoff]

If it really is a problem with the date (no hardware clock), then you can install fake-hwclock which early on in the boot process sets the date to the date of the previous shutdown, which is at least better than 1969. Also, I run rpc-statd in addition to mountd and idmapd. I don't know if that helps.

[ueghio]

Unfortunately it doesn't seem to work.
It's the status
$this->bbcode_second_pass_code('', 'rpc-idmapd.service - NFS v4 ID-Name Mapping Daemon
Loaded: loaded (/usr/lib/systemd/system/rpc-idmapd.service; disabled)
Active: active (running) since Thu 1970-01-01 01:00:28 CET; 44 years 1 months ago
Docs: man:rpc.idmapd(8)
man:idmapd.conf(5)
Main PID: 171 (rpc.idmapd)
CGroup: /system.slice/rpc-idmapd.service
└─171 /usr/bin/rpc.idmapd

Jan 01 01:00:28 pogoplug systemd[1]: Started NFS v4 ID-Name Mapping Daemon.

rpc-mountd.service - NFS Mount Server Daemon
Loaded: loaded (/usr/lib/systemd/system/rpc-mountd.service; disabled)
Active: inactive (dead)
Docs: man:rpc.mountd(8)

Jan 01 01:00:28 pogoplug systemd[1]: Dependency failed for NFS Mount Server Daemon.

rpc-statd.service - NFS v2/3 Network Status Monitor Daemon
Loaded: loaded (/usr/lib/systemd/system/rpc-statd.service; enabled)
Active: active (running) since Thu 1970-01-01 01:00:28 CET; 44 years 1 months ago
Docs: man:rpc.statd(8)
Main PID: 145 (rpc.statd)
CGroup: /system.slice/rpc-statd.service
└─145 /usr/bin/rpc.statd

Jan 01 01:00:27 pogoplug rpc.statd[145]: Version 1.2.9 starting
Jan 01 01:00:27 pogoplug sm-notify[146]: Version 1.2.9 starting
Jan 01 01:00:27 pogoplug rpc.statd[145]: Failed to read /var/lib/nfs/state: Success
Jan 01 01:00:27 pogoplug rpc.statd[145]: Initializing NSM state
Jan 01 01:00:27 pogoplug rpc.statd[145]: Running as root. chown /var/lib/nfs to choose different user
Jan 01 01:00:28 pogoplug systemd[1]: Started NFS v2/3 Network Status Monitor Daemon.
')
Thank you anyway

[ueghio]

Does anybody know how to use the "sleep" command in fstab? Could it be a workaround?

[yeehaa]

maybe start the nfs services using a script after boot is complete? i have the same issue on a newly setup pogo E02.. nfs works fine on my goflex home.. but for the pogo there is something going on with the systemclock update at boot..