Tim, If you use that method your NFS share properties will only live in the hosts dfstab file. By setting zfs properties you achieve the same thing except the share follows the zfs filesystem. (nas/vol0 in this example)
e.g. zfs set sharenfs=rw,nosuid,root=vmware-host-ip-etc sp1/nas/vol0 zfs set mountpoint=/export/ss1-nas-vol0 sp1/nas/vol0 If you do a zfs send to an external zfspool the property lives on it without a tab change. Warning: rw=all = danger! Do you really want every node to have rw access? -----Original Message----- From: Tim Dunphy [mailto:bluethu...@gmail.com] Sent: Monday, May 07, 2012 4:12 PM To: Discussion list for OpenIndiana Subject: Re: [OpenIndiana-discuss] unable to start nfs well, shiver me timbers!!! all I had to do to get the service started was to add an entry in dfstab: root@openindiana:/export/home/bluethundr# tail /etc/dfs/dfstab# Do not modify this file directly. # Use the sharemgr(1m) command for all share management # This file is reconstructed and only maintained for backward # compatibility. Configuration lines could be lost. # # share [-F fstype] [ -o options] [-d "<text>"] <pathname> [resource] # .e.g, # share -F nfs -o rw=engineering -d "home dirs" /export/home2 share -F nfs -o sec=sys,rw=all -d "xen share" /tank/xen I had the idea of adding the entry and starting the service. So glad that did the trick! Thanks for your help guys! On Mon, May 7, 2012 at 3:27 PM, Tim Dunphy <bluethu...@gmail.com> wrote: >> To be pedantic, you don't share a zpool, but filesystem directories, >> over NFS :) > > Point taken. :) > >> In fact, you should check the output of "svcs -a | grep nfs" for >> hints whether your nfs/server is "online" or in another state; if it >> is not online, you can drill down to see why. > > It definitely looks as it's offline at the moment. > > root@openindiana:/export/home/bluethundr# svcs -a | grep > svc:/network/nfs/server:default disabled 14:33:21 > svc:/network/nfs/server:default > > Forgive my ignorance but I'm a little new at solaris. The only > drilling down I've done so far is this: > > svc:/network/nfs/server:default (NFS server) > State: disabled since May 7, 2012 03:11:28 PM EDT > Reason: Temporarily disabled by an administrator. > See: http://illumos.org/msg/SMF-8000-1S > See: nfsd(1M) > See: /var/svc/log/network-nfs-server:default.log > Impact: This service is not running. > > root@openindiana:~# svcs -l nfs/server fmri > svc:/network/nfs/server:default name NFS server enabled > false (temporary) state disabled next_state none state_time > May 7, 2012 03:11:28 PM EDT logfile > /var/svc/log/network-nfs-server:default.log > restarter svc:/system/svc/restarter:default contract_id dependency > require_any/error svc:/milestone/network (online) dependency > require_all/error svc:/network/nfs/nlockmgr (online) dependency > optional_all/error svc:/network/nfs/mapid (online) dependency > require_all/restart svc:/network/rpc/bind (online) dependency > optional_all/none svc:/network/rpc/keyserv (online) dependency > optional_all/none svc:/network/rpc/gss (online) dependency > optional_all/none svc:/network/shares/group (multiple) dependency > optional_all/none svc:/system/filesystem/reparse (online) dependency > require_all/error svc:/system/filesystem/local (online) > > Everything that NFS relies on seems to be 'online' and the only > difference is in the entry for vc:/network/shares/group (multiple). > Which seems to correspond to the message I'm getting when I try to > start the service recursively: > > > root@openindiana:~# svcadm enable -r nfs/server > svcadm: svc:/milestone/network depends on svc:/network/physical, which > has multiple instances. > > >> What happens if you do this? >> svcadm enable nfs/server > > If I do this it's basically the same thing as before: > > root@openindiana:~# svcadm enable nfs/server root@openindiana:~# svcs > -a | grep nfs | grep server disabled 15:17:31 > svc:/network/nfs/server:default > >>Do you have any NFS shares defined via zfs sharenfs attribute and/or >>/etc/dfs/dfstab file (legacy)? What does the "share" >>command report (should return a list of defined shares, if any), >>similar to "exportfs" output. > > root@openindiana:~# share > root@openindiana:~# > > root@openindiana:~# cat /etc/dfs/dfstab > > # Do not modify this file directly. > # Use the sharemgr(1m) command for all share management # This file is > reconstructed and only maintained for backward # compatibility. > Configuration lines could be lost. > # > # share [-F fstype] [ -o options] [-d "<text>"] <pathname> > [resource] # .e.g, # share -F nfs -o rw=engineering -d > "home dirs" /export/home2 > > This is a brand-new setup and I haven't attempted to setup any shares > prior to trying to setup NFS. > > > Basically my goal is to share the top directory of my zpool which > looks like this: > > root@openindiana:~# df -h /tank > Filesystem size used avail capacity Mounted on tank > > 1.3T 44K 1.3T 1% /tank > > "tank" is in reality a super fast raidz2 disk array consisting of 4 > very fast hard drives. > > Thank you all for your input. > > Best, > Tim > > > > > > > > > > On Mon, May 7, 2012 at 2:42 PM, Jim Klimov <jimkli...@cos.ru> wrote: >> 2012-05-07 22:04, Tim Dunphy wrote: >>> >>> Hello openindiana, >>> >>> I appreciate the help you've provided so far. This OI box is >>> turning out completely awesome! >> >> >> Yup! ;) >> >> >>> Currently I am trying to start an nfs server so that I can share my >>> zpool across the network. >> >> >> To be pedantic, you don't share a zpool, but filesystem directories, >> over NFS :) In case of ZFS-NFS integration, you share filesystem >> datasets at their roots (and their child filesystem datasets by >> default - unless you override the sharenfs attribute in the children). >> >> >>> >>> However when I attempt to start the service I am getting an error: >>> >>> root@openindiana:~# svcadm enable -r nfs/server >>> svcadm: svc:/milestone/network depends on svc:/network/physical, >>> which has multiple instances. >>> >>> >>> Now, I do have two physical links that I am currently using to >>> create >> >> >> >> As James pointed out, this error message is not what you interpreted >> it to be. Moreover, I often see it (in the same situation) on other >> Solaris 10+ releases, and it is in fact just a warning from SMF to >> make you notice that it is possible that the service you want might >> rely on a different service than you have configured, or somesuch. >> >> In fact, you should check the output of "svcs -a | grep nfs" for >> hints whether your nfs/server is "online" or in another state; if it >> is not online, you can drill down to see why. >> >> HTH, >> //Jim >> >> >> _______________________________________________ >> OpenIndiana-discuss mailing list >> OpenIndiana-discuss@openindiana.org >> http://openindiana.org/mailman/listinfo/openindiana-discuss > > > > -- > GPG me!! > > gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B -- GPG me!! gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B _______________________________________________ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss _______________________________________________ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
