Subcommands

With no subcommand entered, a sharemgr command with the -h option displays a usage message for all subcommands.

The following subcommands follow sharemgr on a command line. Commands take the form:

% sharemgr <subcommand> [options]

create [-nvh] [-P proto [-p property=value]] group

Create a new group with specified name.

If -n is specified, the command checks only the validity of the command and that the group does not already exist.

If no protocol is specified, all known protocols are enabled for the specified group. If a protocol is specified, only that protocol is enabled. You can specify properties for a specified protocol.

If group exists, use of -P adds the specified protocol to that group.

As an example of the create subcommand, the following command creates a new group with the name mygroup.

# sharemgr create mygroup

Because no protocol was specified in the preceding command, all defined protocols will be enabled on the group.

delete [-nvh] [-P proto] [-f] group

Delete the specified group. If the group is not empty, you can use the -f option to force the deletion, which unshares and removes all shares from the group before removing the group itself.

If you specify a protocol, rather than deleting the whole group, this subcommand deletes the protocol from the group.

The -n option can be used to test the syntax of the command.

As an example, the following command removes the group mygroup from the configuration if it is empty.

# sharemgr delete mygroup

The following command removes any existing shares prior to removing the group.

# sharemgr delete -f mygroup

Note the use of the force (-f) option, above.

list [-vh] [-P proto]

List the defined groups.

If a protocol is specified, list only those groups that have the specified protocol defined.

If the verbose option is specified, the current state of the group and all protocols enabled on the group are listed as well. For example:

# sharemgr list -v
mygroup    enabled    nfs
rdonlygrp  disabled   nfs

show [-pvxh] [-P proto] [group...]

Shows the contents of the specified group(s).

If the verbose option is specified, the resource name and description of each share is displayed if they are defined. Otherwise, only the share paths are displayed. Also, when temporary shares are listed, they are prefixed with an asterisk (‘*’).

If the -p option is specified, all options defined for the protocols of the group are displayed, in addition to the display without options. If the -P option is used, the output is limited to those groups that have the specified protocol enabled. If the -x option is specified, output is in XML format and the -p and -v options are ignored, because all information is included in the XML.

The following example illustrates the use of the -p option.

# sharemgr show -p mygroup
default nfs=()
    * /data/backup
mygroup nfs=(nosuid=true)
      /export/home/home0
      /export/home/home1

The following example illustrates the use of the -v option.

# sharemgr show -v mygroup
mygroup
    HOME0=/export/home/home0    "Home directory set 0"
    HOME1=/export/home/home1    "Home directory set 1"

ZFS managed shares are handled in a way similar to the way NFS shares are handled. These shares appear as subgroups within the parent group zfs. The subgroups are always prefixed with zfs/ and use the ZFS dataset name for the rest of the name. The mount point and any sub-mounts that inherit sharing are shown as the shares of the subgroup. For example:

# sharemgr show -vp zfs
zfs        nfs=()
    zfs/ztest
          /ztest
          /ztest/backups

set [-nvh] -P proto [-S optionset] [-p property= value]... [-s sharepath] group

Set protocol-specific properties on the specified group.

The -P option is required and must specify a valid protocol.

Optionsets are protocol-specific sets of properties that can be negotiated by the protocol client. For NFS, optionsets are equivalent to security modes as defined in nfssec(7). If -S optionset is specified, the properties are applied to the selected optionset. Otherwise they are applied to the general optionset.

Together, -P and -S select a specific view of the group's options on which to work.

Property values are strings. A specified property is set to a new value if the property already exists or is added to the protocol if it does not already exist.

In the general case, at least one property must be set. If -S is specified, properties can be omitted and the specified optionset is enabled for the protocol.

The -s option allows setting properties on a per-share basis. While this is supported, it should be limited to managing legacy shares and to the occasional need for an override of a group-level property or placing an additional property on one share within a group.

An example of this subcommand:

# sharemgr set -P nfs -p anon=1234 mygroup

The preceding command adds the property anon=1234 to the nfs view of group mygroup. If mygroup has existing shares, they will all be reshared with the new property value(s).

unset [-nvh] -P proto [-S optionset] [-p property]... [-s sharepath] group

Unset the specified properties for the protocol or for the specified optionset of the protocol.

In the general case, at least one property must be set. If -S is specified, properties can be omitted and the specified optionset is removed from the protocol.

The -s option allows removing a share-specific property.

An example of this subcommand:

# sharemgr unset -P nfs -p anon mygroup

The preceding command removes the anon= property from the nfs view of group mygroup. If mygroup has existing shares, they will all be reshared with the new property value(s).

add-share [-nth] [-r resource-name] [-d "description text"] -s sharepath group

Add a new share to the specified group.

The -s option is mandatory and takes a full directory path.

If either or both of -d and -r are specified, they specify values associated with the share. -d provides a description string to document the share and -r provides a protocol-independent resource name. Resource names are not used by NFS at this time but can be specified. These names currently follow the same naming rules as group names.

The temporary option (-t) results in the share being shared but not stored in the configuration repository. This option is intended for shares that should not survive a reboot or server restart, or for testing purposes. Temporary shares are indicated in the show subcommand output with an asterisk (‘*’) preceding the share.

If sharepath is a ZFS path and that path is added to the zfs group, sharemgr creates a new ZFS subgroup; the new share is added to that subgroup. Any ZFS sub-filesystems under the ZFS filesystem designated by sharepath will inherit the shared status of sharepath.

The effect of the add-share subcommand on a ZFS dataset is determined by the values of the sharesmb and sharenfs properties of that dataset.

See zfs(8) for a description of the sharesmb and sharenfs properties.

The following are examples of the add-share subcommand.

# sharemgr add-share -s /export/home/home0 -d "home \
directory set 0" -r HOME0 mygroup

# sharemgr add-share -s /export/home/home1 -d "home \
directory set 1" -r HOME1 mygroup

The preceding commands add /export/home/home0 and /export/home/home1 to the group mygroup. A descriptive comment and a resource name are included.

move-share [-nvh] -s sharepath destination-group

Move the specified share from the group it is currently in to the specified destination group. The move-share subcommand does not create a group. A specified group must exist for the command to succeed.

The following is an example of this subcommand.

# sharemgr move-share -s /export/home/home1 newgroup

Assuming /export/home/home1 is in the group mygroup, the preceding command moves /export/home/home1 to the group newgroup and unshares and then reshares the directory with the properties associated with newgroup.

remove-share [-fnvh] -s sharepath group

Remove the specified share from the specified group. The force (-f) option forces the share to be removed even if it is busy.

You must specify the full path for sharepath. For group, use the subgroup as displayed in the output of the sharemgr show command. Note that if there are subshares that were created by inheritance, these will be removed, along with the parent shares.

set-share [-nvh] [-r -resource] [-d "description text"] -s sharepath group

Set or change the specified share's description and resource values. One use of set-share is to rename a resource. The syntax for this use of the subcommand is:

# sharemgr set-share -r current_name=new_name -s sharepath group

enable [-nvh] [group... | -a]

Enable the specified group(s), or (with -a) all groups, and start sharing the contained shares. This state persists across reboots.

An enabled group will be shared whenever the corresponding SMF service instance is enabled. sharemgr will start the SMF service instance if it is not currently online.

disable [-nvh] [group... | -a]

Disable the specified group(s), or (with -a) all groups, and unshare the shares that they contain. This state persists across reboots.

A disabled group will not be shared even if the corresponding SMF service instance is online. This feature is useful when you do not want a group of shares to be started at boot time.

start [-vh] [-P proto] [group... | -a]

Start the specified group, or (with -a) all groups. The start subcommand is similar to enable in that all shares are started, but start works only on groups that are enabled. start is used by the SMF to start sharing at system boot.

A group will not start sharing if it is in the sharemgr disabled state. However, the corresponding SMF service instance will be started.

Note that the start subcommand is similar to the shareall(8) command in that it starts up only the configured shares. That is, the enabled shares will start being shared, but the configuration state is left the same. The command:

# sharemgr start -a

...is equivalent to:

# shareall

stop [-vh] [-P proto] [group... | -a]

Stop the specified group, or (with -a) all groups. The stop subcommand is similar to disable in that all shares are no longer shared, but it works only on groups that are enabled. stop is used by the SMF to stop sharing at system shutdown.

Note that the stop subcommand is similar to the unshareall(8) command in that all active shares are unshared, but the configuration is left the same. That is, the shares are stopped but the service instances are left enabled. The command:

# sharemgr stop -a

...is equivalent to:

# unshareall

share [-F fstype] [-p] [-o optionlist] [-d descriptoion] [pathname [resoucename]]

Shares the specified path in the default share group. This subcommand implements the share(8) functionality. Shares that are shared in this manner will be transient shares. Use of the -p option causes the shares to be persistent.

unshare [-F fstype] [-p] [-o optionlist] sharepath

Unshares the specified share. This subcommand implements the unshare(8) functionality. By default, the unshare is temporary. The -p option is provided to remove the share from the configuration in a way that persists across reboots.

Supported Properties

Properties are protocol-specific. Currently, only the NFS and SMB protocols are supported. Properties have the following characteristics:

• Values of type boolean take either true or false.
• Values of type value take a numeric value.
• Values of type file take a file name and not a file path.
• Values of type access-list are described in detail following the descriptions of the NFS properties.

The general properties supported for NFS are:

abe=boolean

Set the access-based enumeration (ABE) policy for a share. When set to ‘true’, ABE filtering is enabled on this share and directory entries to which the requesting user has no access will be omitted from directory listings returned to the client. When set to ‘false’ or not defined, ABE filtering will not be performed on this share. This property is not defined by default.

false

Disable ABE for this share.

true

Enable ABE for this share.

aclok=boolean

Allows the NFS server to do access control for NFS Version 2 clients (running SunOS 2.4 or earlier). When aclok is set on the server, maximum access is given to all clients. For example, with aclok set, if anyone has read permissions, then everyone does. If aclok is not set, minimum access is given to all clients.

ad-container

Specifies the AD container in which to publish shares.

The AD container is specified as a comma-separated list of attribute name-value pairs using the LDAP distinguished name (DN) or relative distinguished name (RDN) format. The DN or RDN must be specified in LDAP format using the cn=, ou=, and dc= prefixes:

• cn represents the common name
• ou represents the organizational unit
• dc represents the domain component

cn=, ou=, and dc= are attribute types. The attribute type used to describe an object's RDN is called the naming attribute, which, for ADS, includes the following object classes:

• cn for the user object class
• ou for the organizational unit (OU) object class
• dc for the domainDns object class

anon=uid

Set uid to be the effective user ID of unknown users. By default, unknown users are given the effective user ID UID_NOBODY. If uid is set to -1, access is denied.

catia=boolean

CATIA V4 uses characters in file names that are considered to be invalid by Windows. CATIA V5 is available on Windows. A CATIA V4 file could be inaccessible to Windows clients if the file name contains any of the characters that are considered illegal in Windows. By default, CATIA character substitution is not performed. If the catia property is set to true, the following character substitution is applied to file names.

CATIA    CATIA
V4 UNIX  V5 Windows
  "      \250   0x00a8  Dieresis
  *      \244   0x00a4  Currency Sign
  /      \370   0x00f8  Latin Small Letter O with Stroke
  :      \367   0x00f7  Division Sign
  <      \253   0x00ab  Left-Pointing Double Angle Quotation Mark
  >      \273   0x00bb  Right-Pointing Double Angle Quotation Mark
  ?      \277   0x00bf  Inverted Question Mark
  \      \377   0x00ff  Latin Small Letter Y with Dieresis
  |      \246   0x00a6  Broken Bar

cksum=cksumlist

Set the share to attempt to use end-to-end checksums. The value cksumlist specifies the checksum algorithms that should be used.

csc=value

Set the client-side caching policy for a share. Client-side caching is a client feature and offline files are managed entirely by the clients.

The following are valid values for the csc property:

manual

Clients are permitted to cache files from the specified share for offline use as requested by users. However, automatic file-by-file reintegration is not permitted. manual is the default value.

auto

Clients are permitted to automatically cache files from the specified share for offline use and file-by-file reintegration is permitted.

vdo

Clients are permitted to automatically cache files from the specified share for offline use, file-by-file reintegration is permitted, and clients are permitted to work from their local cache even while offline.

disabled

Client-side caching is not permitted for this share.

guestok=boolean

Set the guest access policy for the share. When set to true guest access is allowed on this share. When set to false or not defined guest access is not allowed on this share. This property is not defined by default.

An idmap(8) name-based rule can be used to map guest to any local username, such as guest or nobody. If the local account has a password in /var/smb/smbpasswd the guest connection will be authenticated against that password. Any connection made using an account that maps to the local guest account will be treated as a guest connection.

Example name-based rule:

# idmap add winname:Guest unixuser:guest

index=file

Load file rather than a listing of the directory containing this file when the directory is referenced by an NFS URL.

log=tag

Enables NFS server logging for the specified system. The optional tag determines the location of the related log files. The tag is defined in /etc/nfs/nfslog.conf. If no tag is specified, the default values associated with the global tag in /etc/nfs/nfslog.conf is used. Support of NFS server logging is available only for NFS Version 2 and Version 3 requests.

nosub=boolean

Prevents clients from mounting subdirectories of shared directories. For example, if /export is shared with the nosub option on server ‘wool’ then an NFS client cannot do:

# mount -F nfs wool:/export/home/mnt

NFS Version 4 does not use the MOUNT protocol. The nosub option applies only to NFS Version 2 and Version 3 requests.

nosuid=boolean

By default, clients are allowed to create files on a shared file system with the setuid or setgid mode enabled. Specifying nosuid causes the server file system to silently ignore any attempt to enable the setuid or setgid mode bits.

public=boolean

Moves the location of the public file handle from root (/) to the exported directory for WebNFS-enabled browsers and clients. This option does not enable WebNFS service; WebNFS is always on. Only one file system per server can have the public property. You can apply the public property only to a share and not to a group.

NFS also supports negotiated optionsets for supported security modes. The security modes are documented in nfssec(7). The properties supported for these optionsets are:

charset=access-list

Where charset is one of: euc-cn, euc-jp, euc-jpms, euc-kr, euc-tw, iso8859-1, iso8859-2, iso8859-5, iso8859-6, iso8859-7, iso8859-8, iso8859-9, iso8859-13, iso8859-15, koi8-r.

Clients that match the access-list for one of these properties will be assumed to be using that character set and file and path names will be converted to UTF-8 for the server.

ro=access-list

Sharing is read-only to the clients listed in access-list; overrides the rw suboption for the clients specified. See the description of access-list below.

rw=access-list

Sharing is read-write to the clients listed in access-list; overrides the ro suboption for the clients specified. See the description of access-list below.

none=access-list

Access is not allowed to any client that matches the access list. The exception is when the access list is an asterisk (‘*’), in which case ro or rw can override none.

root=access-list

Only root users from the hosts specified in access-list have root access. See details on access-list below. By default, no host has root access, so root users are mapped to an anonymous user ID (see the anon=uid option described above). Netgroups can be used if the file system shared is using UNIX authentication (AUTH_SYS).

root_mapping=uid

For a client that is allowed root access, map the root UID to the specified user id.

window=value

When sharing with sec=dh (see nfssec(7)), set the maximum lifetime (in seconds) of the RPC request's credential (in the authentication header) that the NFS server allows. If a credential arrives with a lifetime larger than what is allowed, the NFS server rejects the request. The default value is 30000 seconds (8.3 hours). This property is ignored for security modes other than dh.

The general properties supported for SMB are:

encrypt=string

Controls SMB3 per-share encryption. This is similar to the global smbd/encrypt option. For requests on a particular share, the server's behavior is controlled by the stricter of this option and smbd/encrypt.

When set to disabled, the server will not ask clients to encrypt requests. When set to enabled, the server will ask clients to encrypt requests, but will not require that they do so. Any message than can be encrypted will be encrypted. When set to required, the server will deny access to or disconnect any client that does not support encryption or fails to encrypt requests that they should.

In other words, the enabled behavior is that any message that CAN be encrypted SHOULD be encrypted, while the required behavior is that any message that CAN be encrypted MUST be encrypted.

This property is not defined by default.

ro=access-list

Sharing is read-only to the clients listed in access-list; overrides the rw suboption for the clients specified. See the description of access-list below.

rw=access-list

Sharing is read-write to the clients listed in access-list; overrides the ro suboption for the clients specified. See the description of access-list below.

none=access-list

Access is not allowed to any client that matches the access list. The exception is when the access list is an asterisk (‘*’), in which case ro or rw can override none.

Access List Argument

The access-list argument is either the string ‘*’ to represent all hosts or a colon-separated list whose components can be any number of the following:

hostname

The name of a host. With a server configured for DNS or LDAP naming in the nsswitch.conf(5) ‘hosts’ entry, a hostname must be represented as a fully qualified DNS or LDAP name.

netgroup

A netgroup contains a number of hostnames. With a server configured for DNS or LDAP naming in the nsswitch.conf(5) ‘hosts’ entry, any hostname in a netgroup must be represented as a fully qualified DNS or LDAP name.

domainname.suffix

To use domain membership the server must use DNS or LDAP, rather than, for example, NIS, to resolve hostnames to IP addresses. That is, the hosts entry in the nsswitch.conf(5) must specify ‘dns’ or ‘ldap’ ahead of ‘nis’, because only DNS and LDAP return the full domain name of the host. Other name services, such as NIS, cannot be used to resolve hostnames on the server because, when mapping an IP address to a hostname, they do not return domain information. For example, for the IP address 172.16.45.9:

NIS

Returns: myhost

DNS or LDAP

Returns: myhost.mydomain.example.com

The domain name suffix is distinguished from hostnames and netgroups by a prefixed dot. For example:

rw=.mydomain.example.com

A single dot can be used to match a hostname with no suffix. For example, the specification:

rw=.

...matches ‘mydomain’ but not ‘mydomain.example.com’. This feature can be used to match hosts resolved through NIS rather than DNS and LDAP.

network

The network or subnet component is preceded by an at-sign (‘@’). It can be either a name or a dotted address. If a name, it is converted to a dotted address by getnetbyname(3SOCKET). For example:

=@mynet

...is equivalent to:

=@172.16 or =@172.16.0.0

The network prefix assumes an octet-aligned netmask determined from the zeroth octet in the low-order part of the address up to and including the high-order octet, if you want to specify a single IP address. In the case where network prefixes are not byte-aligned, the syntax allows a mask length to be specified explicitly following a slash (‘/’) delimiter. For example:

=@theothernet/17 or =@172.16.132/22

...where the mask is the number of leftmost contiguous significant bits in the corresponding IP address.

A prefixed minus sign (‘-’) denies access to a component of access-list. The list is searched sequentially until a match is found that either grants or denies access, or until the end of the list is reached. For example, if host ‘terra’ is in the netgroup ‘engineering’, then:

rw=-terra:engineering

...denies access to ‘terra’, but:

rw=engineering:-terra

...grants access to ‘terra’.