 |
You can use the contents
of the Figure 1 attribute table file as a
NFS server sample. The attributes table can be found as GFSAPATT in
the NFSSAMP library.
Figure 1. NFS
system server sample attribute table#######################################################################
# #
# Z/OS Network File System Server Sample Attribute Table @L6C #
# #
# PROPRIETARY STATEMENT= #
# LICENSED MATERIALS - PROPERTY OF IBM #
# THIS MODULE IS "RESTRICTED MATERIALS OF IBM" #
# 5650-ZOS @017C #
# COPYRIGHT IBM CORPORATION 1991, 2012 @017C #
# SEE IBM COPYRIGHT INSTRUCTIONS #
# END PROPRIETARY STATEMENT #
# Copyright IBM Corp. 1991, 2011 @P0GC #
# Copyright SUN Microsystems, Inc & #
# Electronic Data Systems Corp. 1988, 1989 #
# #
#######################################################################
# #
# change activities: #
# 1. 5/10/91 - Release it for MVS/DFP V3 #
# 2. 1/30/92 - Updates for VSCR #
# 3. 8/06/92 - Change mintimeout default #
# 4. 8/31/92 - Add PCNFSD #
# 5. 9/20/92 - R2 updates #
# $L3X=NFS,HDZ11NP,931230,SJPLJST: Change NFSTASKS default and @L3XA#
# add XLAT keyword @L3XA#
# $L3L=KA90033,HDZ11NP,940405,SJPLJST: Add RESTIMEOUT keyword @L3LA#
# #
# $L33=NFS,HDZ11NP,940613,SJPLTEM: Add SMF keyword @L33A#
# $01=OW12199,HDZ11NP,950323,SJPLTEM: Add HFS keywork @01A#
# $P1=KA00045,HDZ11SP,960111,SJPLTEM: Updates for DFSMS 1.3 @P1A#
# $P2=KA00107,HDZ11SM,960415,SJPLTEM: Remove MODEL attribute @P2A#
# $L59=NFS,HDZ11TS,970226,SJPLBPF: File Ext. Support @L59A#
# $P3=KAB0033,HDZ11TS,970701,SJPLPKL: Add # comment char after @P3A#
# xlat keyword @P3A#
# $P4=KAB0114,HDZ11TS,971030,SJPLTEM: Chg DFSMS/MVS->OS/390 @P4A#
# $L53=NFS,HDZ11TS,971031,SJPLBPF: WebNFS Support @L53A#
# $P5=KAB0379,HDZ11TS,980512,SJPLBPF: Default changes @P5A#
# $L5D=NFS,HDZ11TS,980821,SJPLBPF: NC Support OW34846=@L5DA#
# $L5X=NFS,HDZ11TS,980820,SJPLTEM: Filename delimiter OW34846=@L5XA#
# $LA1=OW38745,HDZ11TS,981209,SJPLRMS: nfstasks(n,m,o) @LA1A#
# $02=OW40268,HDZ11TS,990727,SJPLRMS: Comment out nfstasks flag @02A#
# $03=OW42036,HDZ11TS,991213,SJPLBPF: New readdirtimeout keyword @03A#
# $04=OW43829,HDZ11TS,000410,SJPLBPF: Lower readdirtimeout limit @04A#
# $LA7=OW46949,HDZ11TS,000921,IBSKEK: TEXT/BINARY on a single @LA7A#
# mount point @LA7A#
# $P6=KAD0016,HDZ11TS,001106,SJPLBPF: Allow up to 100 hfs tasks @P6A#
# $05=OW48939,HDZ11TS,010415,SJPLJST: rddr cookie verifier @05A#
# $LC1=OW49104,HDZ11TS,010115,IBSNIV: File Tagging Support @LC1A#
# $06=OW51358,HDZ11TS,010921,IBSPIV : New hfsfbtimeout keyword @06A#
# $07=OW54351,HDZ11TS,020422,IBSKVV : New upcase and @07A#
# mixcase keywords @07A#
# $08=OW55830,HDZ11TS,020819,IBSKVV :SMF activate at NFSS startup @08A#
# $L6=NFS,HDZ11US,030405,SJPLMB: Changed OS/390 to Z/OS @L6A#
# $L66=NFS,HDZ11US,030303,IBSMVB: NFSS 878 abend handling @L66A#
# $LCE=OW55734,HDZ11TS,020701,IBSNIV: File Tagging Support @LCEA#
# $L74=NFS,HDZ11VS,031015,SJPLJST: NFS ver 4 protocol support @L74A#
# $09=OA03523,HDZ11TS,030515,IBSVKR: Extend RETRIEVE attr to HFS @09A#
# $0A=OA05684,HDZ11TS,031208,IBSNIV: New remount/noremount keyword@0AA#
# $L76=NFS,HDZ11VS,040119,IBSDYP: NFS Server DHCP Support @L76A#
# $L74=NFS,HDZ11VS,040322,SJPLSLH: Added MVSMNT PRocessing Attr @L74BA#
# $L74=NFS,HDZ11VS,040805,SJPLMB: Added DENYRW/NODENYRW Attrs @L74M5A#
# $L7E=NFS,HDZ11VS,041214,SJPLMB: @L7EA#
# Legal Rqmt: Change "OPEN EDITION" to "z/OS UNIX" @L7EA#
# $P07=KAJ0262,HDZ11VS,050204,SJPLMB: @P07A#
# 1. Add missing comment delimiters @P07A#
# 2. Move MVSMNT to Processing Attributes Section @P07A#
# 2. Remove nohfs option. It is no longer supported. @P07A#
# $0B=OA08867,HDZ11US,040929,IBSKYL: Correct typo for @L66A @0BA#
# $P08=KAJ0243,HDZ11VS,050331,SJPLSCA: 64bit fileid support @P08A#
# $0C=OA12850,HDZ118N,050811,SJPLSLH: MVSMNT and SAF text added @0CA#
# $0D=OA14044,HDZ118N,051208,IBSNIV: Correct nfstasks(n, ,o) @0DA#
# $L74=NFS,HDZ11VS,041020,SJPLKU: @L74AA#
# Added mvssec(),hfssec() and pubsec(). @L74AA#
# $L81=NFS,HDZ118N,050908,IBSDYP: Exports file netgroup support @L81A#
# $L82=NFS,HDZ118N,050808,SJPLJFA: Stringprep Support @L82A#
# $0E=OA12994,HDZ118N,060106,SJPLRAS: @0EA#
# Added nordrcache @0EA#
# $P09=K9N0086,HDZ119N,060807,SJPLMB: @P09A#
# 1. Added NFS v4 attr/read/write timeout vs lease @P09A#
# time restriction note. @P09A#
# $P0A=K9N0099,HDZ119N,060823,SJPLMB: @P0AA#
# 1. Changed nfstasks note wording slightly. @P0AA#
# $P0B=K9N0317,HDZ1AN0,070315,SJPLRAS: @P0BA#
# 1. Changed comment for SETOWNERNOBODY attribute. @P0BA#
# $0F=OA20232,HDZ1AN0,070321,IBSNIV: @0FA#
# 1. New second parameter of writetimeout attribute added@0FA#
# $LAG=NFS,HDZ119N,070410,IBSDYP: @LAGA#
# 1. Added NFSv4domain attribute for NFSv4 name mapping @LAGA#
# $P0C=KN00022,HDZ1AN0,070615,IBSDYP: @P0CA#
# 1. Changed comment for NFSV4DOMAIN attribute. @P0CA#
# $P0D=KN00025,HDZ1AN0,070704,SJPLJST: # TCP/UDP tasks spec @P0DA#
# $LAH=NFS,HDZ1AN0,070831,SJPLMB: @LAHA#
# 1. Added ZNFSClient attribute for z/OS NFS Client @LAHA#
# $P0E=KN00144,HDZ1AN0,071024,IBSDYP: multi-domain support for @P0EA#
# NFSv4 name mapping @P0EA#
#-------------------- Release V1r11 -----------------------------@LCQC#
# $LBB=NFS,HDZ1B1N,080404,SJPLMB: @LBBA#
# 1. Added MVSPREFIX and IMPPREFIX site attributes for @LBBA#
# the extended path prefix support. @LBBA#
# $LBK=__NFS__,HDZ1B1N,080414,IBSVAE: @LBKA#
# 1. Delay Detection @LBKA#
# $P0F=KBN0052,HDZ1B1N,080814,IBSVAE: @P0FA#
# 1. Correct wording in description of DlyDTimeout @P0FA#
# $LBC=__NFS__,HDZ1B1N,080818,SJPLJST: @LBCA#
# 1. Add DELEGATION/NODELEGATION Site Attributes @LBCA#
# $010=OA24375,HDZ1B1N,080605,IBSVKR: @010A#
# 1. h3_rddr enhancement @010A#
#-------------------- Release V1r12 -----------------------------@LCQA#
# $LCQ=__NFS__,HDZ1C1N,090420,IBSAYG: @LCQA#
# 1. Cache monitoring & reporting - Bufhigh attribute @LCQA#
# now contains 2 values. @LCQA#
# $LCO=__NFS__,HDZ1C1N,090407,IBSVKR : @LCOA#
# 1. Add new operands in the site attribute: SMF @LCOA#
# $011=OA29577,HDZ1C1N,090720,SJPLMM: @011A#
# 1. Change MAXWINDOW from 256 to 2048 so zNFSS can @011A#
# cache up to 64MB out_of_sequence 32KB_rpc_write. @011A#
# $012=OA27743,HDZ1C1N,090522,IBSVKR: @012A#
# 1. New site attribute for GID inheritance @012A#
# $013=OA24594,HDZ1C1N,080521,IBSNIV: @013A#
# 1. The overflow attribute support. @013A#
# $014=OA31051,HDZ1C1N,091111,IBSNIV: @014A#
# 1. AE for OA24594: Add pound sign to fiche flag. @014A#
# $015=OA30333,HDZ1C1N,091221,IBSSAN: @015A#
# 1. Changed comment for NFSV4DOMAIN attribute. @015A#
#-------------------- Release V1r13 -----------------------------@P0GA#
# $P0G=KCN0160,HDZ1D1N,100325,SJPLRAS: @P0GA#
# 1. bufhigh - restore Server may adjust initial value on @P0GA#
# startup. And change default to 32M from 64M @P0GA#
# $P0H=KCN0164,HDZ1D1N,100402,SJPLRAS: @P0HA#
# 1. bufhigh - restore comment delimiter in front of fiche @P0HA#
# $P0I=KCN0145,HDZ1D1N,100407,SJPLRAS: @P0IA#
# - General Comment Cleanup @P0IA#
# 1. Comment update: nfstasks attribute @P0IA#
# 2. Comment update: mintasks attribute @P0IA#
# 3. Fix Prolog comment @P0IA#
# 4. Comment update: MVSMNT @P0IA#
# 5. Comment update: Added commented out sidefile(dsname) @P0IA#
# 6. Convert lowercase BUFHIGH to bufhigh @P0IA#
# 7. Comment update: RECFM - change "S" to "M" @P0IA#
# $016=OA31545,HDZ1D1N,100311,IBSNIV: @016A#
# 1. The overflow attribute support for V4. @016A#
# (included in R1.12 roll-up APAR OA32835) @016A#
#-------------------- Release V2r1 -----------------------------@017A#
# $017=OA34879,HDZ221N,20111123,IBSNIV: @017A#
# 1. The BLKSIZE for direct access data set. @017A#
# $018=OA35630,HDZ221N,20110216,IBSVAE: @018A#
# 1. symresolve/nosymresolve support. @018A#
# $019=OA36971,HDZ221N,20110701,IBSVKR: @019A#
# 1. New site attribute: mvslogindelay @019A#
# $LEP=__NFS__,HDZ221N,20110926,SJPLHPN: @LEPA#
# 1. Update LOGICALCACHE, BUFHIGH, CACHEWINDOW comments @LEPA#
# with regards to LP64 and the z/OS NFS Server ability @LEPA#
# to utilize the virtual memory above 2GB BAR to @LEPA#
# buffer the logical read/write data to MVS data sets. @LEPA#
# $P0J=KEN0151,HDZ221N,20111012,IBSDYP: @P0JA#
# 1. Move OVERFLOW to Processing Attributes Section. @P0JA#
# $P0K=KEN0169,HDZ221N,20111014,IBSDYP: @P0KA#
# 1. Move OVERFLOW to Processing Attributes Section (PTM @P0KA#
# KEN0151 concerned only PUB updates). @P0KA#
# $P0L=KEN0196,HDZ221N,20111214,SJPLTB: @P0LA#
# 1. Add default value for logicalcache @P0LA#
# #
#######################################################################
#
# This is a prototype site defaults attribute file for the
# Z/OS Network File System Server Sample Attribute Table @L6C
#
# '#' character starts a comment. Comments can appear anywhere.
# White space is ignored when parsing the file.
# Default values are illustrated in the examples in this file
#
# Keywords are not case sensitive. 'BLKS' is the same as 'blks' is
# the same as 'BlKs'.
# All time values are in seconds.
#######################################################################
# The following are known as data set creation attributes. #
#######################################################################
# SPACE specifies the amount of primary and secondary space allocated
# for a new data set. The syntax is:
#
# SPACE(PRIMARY,SECONDARY)
#
# The secondary field is optional (if omitted, the default is taken).
#
# Dimension of allocation is BLKS, TRKS, or CYLS
# RECS is a synonym for BLKS.
space(100,10), blks
# RLSE specifies that unused space should be released from the data
# set the first time that a new data set is closed. For slow clients,
# with long pauses between writes, the RLSE attribute will cause space
# to be released from the primary extent prematurely. Subsequent
# writes will cause secondary space to be allocated.
norlse
# The record format, or RECFM, defines part of the layout of a data
# set: how the records are physically layed out on disk.
#
# Valid RECFM characters are:
#
# V - Variable Length Records (LRECL defines maximum size of
# any record)
# F - Fixed Length Records (LRECL defines the actual length of
# all records)
# U - Undefined Length Records
#
# modified by:
#
# B - Records are Blocked (BLKSIZE defines the size of the block)
# S - Spanned for variable length records
# Standard format for fixed length records
# M - Machine Control Codes
# A - ANSI Control Codes
#
# "A" and "M" are mutually exclusive @P0IC
# "V", "F", and "U" are mutually exclusive
# "S" is not allowed for DSNTYPE(PDS) and DSNTYPE(LIBRARY)
# (refer to DSNTYPE later in this section.)
#
# The BLKSIZE is the size, in bytes, of a physical block on disk.
# BLKSIZE(0) allows the system to choose an optimized block size.
# The system does not determine a block size for Direct Access @017A
# (DA) data sets. z/OS NFS Server uses the following formula @017A
# to calculate the block size for a DA data set depending on @017A
# the record format: @017A
# @017A
# F | FB blksize = lrecl @017A
# V | VB blksize = lrecl + 4 @017A
# VS | VBS blksize = 6160 @017A
#
# LRECL defines the size, in bytes, of a logical record in the data set.
recfm(vb), blksize(0), lrecl(8196)
# The data set organization can be one of:
#
# PS - Physical Sequential
# DA - Direct Access
# INDEXED - VSAM KSDS data set
# NONINDEXED - VSAM ESDS data set
# NUMBERED - VSAM RRDS data set
#
# PS is a good organization for NFS usage, and NONINDEXED is the
# corresponding good VSAM data set for NFS (e.g. with AIX client in
# BINARY mode) usage.
dsorg(ps)
# DSNTYPE specifies whether a PDS or a PDSE is to be created when
# the make directory workstation command is issued.
#
# Valid DSNTYPEs are:
#
# PDS - Create a Partitioned Data Set.
# LIBRARY - Create a Partitioned Data Set Extended.
dsntype(pds)
# Number of Directory blocks for PDS allocation
dir(27)
# The MGMTCLAS specifies the management class associated with the
# file creation.
#
# The syntax is:
#
# mgmtclas(mgmt_class_name)
# The VOLUME (or VOL) attribute enables you to specify the volume
# on which to create the specified data set.
#
# The syntax is:
#
# volume(volser)
# The UNIT attribute enables you to specify the unit on which to
# create the specified data set.
#
# The syntax is:
#
# unit(unit_name)
# The following attributes are used to control VSAM data set
# creation. They are used only if the DSORG parameter defines
# the data set to be type INDEXED, NONINDEXED or NUMBERED.
#
# Refer to appropriate IBM MVS documentation for a more
# complete description of these and other data set creation
# attributes.
# The KEYS(LENGTH,OFFSET) attribute enables you to define the key
# length and offset for a VSAM INDEXED (KSDS) data set. It is used
# only if DSORG is INDEXED.
#
# Valid range for LENGTH is from 1 to 255.
# Valid range for OFFSET is from 0 to 32760.
keys(64,0)
# The RECORDSIZE(AVERAGE,MAXIMUM) attribute enables a user to define
# the average and maximum record sizes for a VSAM data set.
# These two values must be equal for NUMBERED (RRDS) data sets.
#
# Valid range is from 1 to 32760.
recordsize(512,4K)
# The SPANNED and NONSPANNED attributes define whether VSAM
# records will span control intervals. This option does not affect
# non-VSAM variable length record data sets. Use the 'S' option
# with the RECFM attribute for non-VSAM data sets.
nonspanned
# The SHAREOPTIONS attribute defines the cross region and cross
# system file sharing allowed for a VSAM data set.
# Valid range for each argument is from 1 to 4.
shareoptions(1,3)
#######################################################################
# The following are known as processing attributes. #
#######################################################################
# There are three timeout types: attributes, reads and writes.
# The various timeout values are used by the system to determine
# when to close and deallocate an inactive data set after the last
# "attribute", "read", or "write" operation.
# The WRITETIMEOUT is usually kept short, because WRITE
# operations result in exclusive locking, and you'll want to release
# the data set. For slow clients, with long pauses between writes,
# you'll want to increase the WRITETIMEOUT value.
#
# Valid range is from "m" to "n"; where "m" is the argument of
# MINTIMEOUT(m) and "n" is the argument of MAXTIMEOUT(n), unless
# NOMAXTIMEOUT is specified. In that case, "n" is 32767.
#
# xxxxTIMEOUT(n) indicates to deallocate the data set n seconds after
# the "xxxx" operation;
# NOxxxxTIMEOUT indicates not to deallocate the data set after
# the "xxxx" operation;
# Where "xxxx" can be "ATTR", "READ", or "WRITE".
#
# e.g. WRITETIMEOUT(1) indicates to deallocate the data set 1 second
# after a write.
# NOWRITETIMEOUT indicates not to deallocate the data set after
# a write.
# READTIMEOUT(90) indicates to deallocate the data set 90 seconds
# after a read if no further activity against it.
#
# NOTE: When using the NFS version 4 protocol, these timeout @P09A
# values should be set to a value less than or equal to @P09A
# the lease time. Otherwise, it is possible for @P09A
# performance problems to occur when attempting to access @P09A
# MVS data sets. @P09A
# Also, WRITETIMEOUT(n,o) for a data set can be specified. @0FA
# In this case, "o" indicates the number of seconds that NFS @0FA
# Server will wait for data to arrive to complete a partial @0FA
# record before closing the data set. @0FA
# Valid range for "o" is from "n" to "255 * n". @0FA
# The default value for "o" is "4 * n", if not specified. @0FA
# It is preferable to set it as a multiple of "n", otherwise it @0FA
# will be rounded down to the nearest multiple of "n". @0FA
attrtimeout(120), readtimeout(90), writetimeout(30,120) # @0FC
# Processing may be TEXT or BINARY
#
# BINARY is good for using MVS as a disk farm for PCs and AIX machines
# and offers better performance.
#
# TEXT should be specified if it is necessary to share data sets
# containing textual data with other MVS applications.
binary
# @LA7A
# MAPPED should be specified when a mixed set of data types are @LA7A
# to be processed on a single mount point. The determination @LA7A
# of whether the data is to be processed as text or binary @LA7A
# depends on the rules established in the specified side file. @LA7A
# If a file extension cannot be mapped to text or binary, then @LA7A
# the data will be processed according to what has been specified @LA7A
# as binary or text at the mount level, and finally, the site @LA7A
# level. If binary or text is specified at the file level, @LA7A
# the specification overrides the MAPPED specification. @LA7A
# @LA7A
# The syntax is: @LA7A
# @LA7A
# mapped # @LA7A
# @LA7A
# The end of line terminators are:
#
# CR, CRLF, LF, LFCR, or NOEOL.
#
# They define the conversion of records to end-of-line terminators in
# TEXT mode.
# LF should be used for AIX clients;
# CRLF should be used for PC clients.
# (set through client mounts appropriately). When TEXT mode is
# specified, LF is the default.
LF
# BLANKSTRIP or NOBLANKSTRIP affects the processing of trailing
# blanks when reading and writing to Fixed Record data sets with
# text processing enabled.
blankstrip
# The MAPLEADDOT attribute turns on mapping of a file name starting
# with a leading "." from a client to a legal leading "$" for a
# MVS data set name. NOMAPLEADDOT turns off this mapping.
mapleaddot
# The MAPLOWER attribute tells the server to map file names to
# uppercase when received from the client in an NFS request and
# to translate from uppercase to lowercase when returned to the client.
# Keywords are not case sensitive and are unaffected by this option.
#
# The NOMAPLOWER attribute tells the server NOT to do any translation.
# i.e. the server is neither to map uppercase when received from
# the client nor to translate to lowercase when returned to the client.
# All the entries in the EXPORTS file are case sensitive.
# The client MOUNT request must specify the MVS qualifier with
# the correct case to successfully match the EXPORTS file entry.
maplower
# RETRIEVE tells the server to recall a migrated data set on read/write
# access. NORETRIEVE will force the return of "Device not available"
# error for migrated files.
# RETRIEVE can be coded in the following ways:
# RETRIEVE - nowait for recall. @09C
# RETRIEVE(WAIT) - wait for recall.
# RETRIEVE(NOWAIT) - nowait for recall.
# For a quiesced HFS file system: @09A
# RETRIEVE(WAIT) will suspend a request until the file system @09A
# will be unquiesced. @09A
# Others value will force the return of IO error. @09A
# RETRIEVE is the default
retrieve
# The FASTFILESIZE attribute tells the server to calculate approximate
# file sizes from available catalog information and disk geometries.
# This approximate size may cause problems with client applications
# since the size is probably inaccurate. The NOFASTFILESIZE may
# result in decreased performance because the server may read a data
# set, applying the defined processing attributes, to determine the
# exact size of the data set as viewed by an NFS client.
nofastfilesize
# The SETOWNERROOT keyword tells the server to set the user ID in the
# attributes returned to a client for a specified file to 'root' when
# the client is logged on as superuser. SETOWNERNOBODY tells the
# server to set the user ID in the attributes to 'nobody' (65534).@P0BC
setownerroot
# You can have the execute bit for plain files on or off by mount
# point. Turn this option on if you plan to store executables
# or shell scripts on the MVS system on a mount by mount basis.
# It should probably always be off in the site file.
#
# EXECUTEBITON will turn on the execute bits (user, group
# and other) for a mount point's files.
executebitoff
# If the installation intends to customize the translation table, @L3XA
# a new DD card, NFSXLAT, is required in the NFSS startup proc. @L3XA
# @L3XA
# //NFSXLAT DD DSN=dataset_name,DSP=SHR @L3XA
# @L3XA
# Where dataset_name is the name of PDS or PDSE whose @L3XA
# members are the translation tables. @L3XA
# @L3XA
# The XLAT(member) keyword tells the server which member @L3XA
# the server is to use as the installation default translation @L3XA
# table. 'member' is the name of a translation table which @L3XA
# resides in a PDS or PDSE dataset. @L3XA
# @L3XA
# The syntax is: @L3XA
# @L3XA
# xlat(member) # @P3C
# The 'overflow' processing attribute for MVS data path turns on @P0KM
# x37 detection support for PS/PDS MVS data sets for NFS v2/v3/v4 @P0KM
# WRITE operations. It allows ENOSPACE errors to be reported to @P0KM
# NFS Client in a timely manner and to avoid situations @P0KM
# when the z/OS NFS Server closes a data set on timeout expiration@P0KM
# basis with x37 abend which can not be propagated to the NFS @P0KM
# Client. This option may also be activated on the MOUNT level. @P0KM
# Default mode is no overflow detection. @P0KM
# Note: 1. PDSE and VSAM data sets are not supported. @P0KM
# @P0KM
# overflow # @P0KM
# FILEEXTMAP or NOFILEEXTMAP affects the file extension mapping @L59A
# capability. FILEEXTMAP turns on file extension mapping and @L59A
# NOFILEEXTMAP turns it off. This option can be specified at @L59A
# the file command level. The default is NOFILEEXTMAP. @L59A
nofileextmap # @L59A
# SIDEFILE(dsname) specifies the name of the data set that @L59A
# contains the rules for file extension mapping purposes. @L59A
# If a side file name is specified in the attributes data set @L59A
# then it is the default side file for this NFS server. @L59A
# A user can also specify another side file name during a MOUNT @L59A
# operation to be used along with the default. The mapping rules @L59A
# will first be searched in the side file specified during MOUNT @L59A
# and then in the default. To allow file extension mapping a @L59A
# side file name must be specified either as a default or in the @L59A
# MOUNT command. dsname is a fully-qualified MVS data set name @L59A
# without quotation marks. SIDEFILE is only specifiable at the @L59A
# MOUNT level. See GFSAPMAP for sample mapping side file and @L59A
# syntax. @L59A
# sidefile(dsname) # @P0IA
# TAG specifies the newly created files should receive a file tag.@LCEA
# NOTAG specifies the newly created file should be untagged. @LCEA
# The default is NOTAG. @LCEA
notag # @LCEA
# CLN_CCSID(n) specifies the Coded Character Set Identifier(CCSID)@LC1A
# for the remote mounted file system (NFS client) when text is @LC1A
# being translated. @LC1A
# The default (if specified) is 819 (ISO 8859-1 ASCII). @LC1A
cln_ccsid(819) # @LCEC
# SRV_CCSID(n) specifies the Coded Character Set Identifier(CCSID)@LC1A
# for the local file system (Z/OS NFS Server) when @L6C
# a new file is being created. @LC1A
# The default (if specified) is 1047 (Latin Open System EBCDIC). @LC1A
srv_ccsid(1047) # @LCEC
# CONVSERV(technique) - 'technique' specifies technique-search-order
# which Unicode Service will use for specified srv_ccsid(n) and @LCEA
# cln_ccsid(n) code pages. @LCEA
# 'technique' consists of up 5 technique-characters corresponding @LCEA
# to the 5 available techniques: R, E, C, L and M. @LCEA
# The default is CONVSERV(LRE). @LCEA
convserv(LRE) # @LCEA
# MVSMNT specifies that in NFS version 4, a lookup will be @P07M
# treated as if a mount procedure were given for the LOOKUP @P07M
# result object. @P07M
# @P07M
# For any LOOKUPs that do not specify MVSMNT, any Processing @P07M
# Attributess that may have been provided will be merged @P07M
# with any that were in effect for the LOOKUP parent directory. @P07M
# @P07M
# For LOOKUPs that *DO* specify MVSMNT, any other Processing @P07M
# Attributes provided will be merged with the site defaults. @P07M
# MVSMNT cannot be specified for any LOOKUP where the parent @P07M
# directory was navigated to by an mount procedure or a result @P07M
# of an object that was already LOOKUP'ed with MVSMNT. This @P07M
# is to ensure that only a client system mount specifies MVSMNT. @P07M
# If the MVSMNT, or any other, processing attribute is not @P07M
# specified, then SAF checking may be disabled longer than @P07M
# desired due to the z/OS NFS server not being able to detect @P07M
# the end of the mount and the beginning of other access @P07M
# requests. However, access is only allowed to information @P07M
# necessary for the completion of mount related processing. @P07M
# This behavior is required to avoid requiring an mvslogin to @P07M
# be issued prior to mount processing. The EXPORT file may @P07M
# be used to restrict access to file systems, regardless of the @P07M
# specification of any processing attribute. @P07M
# MVSMNT is not to be specified as a site default attribute. @P07M
# There is no default. MVSMNT should not be specified in the @P0IA
# attributes data set. The mvsmnt attribute should only be @P0IA
# specified on NFS Version 4 mount commands. @P0IA
# @P0IA
# NORDRCACHE specifies that the server should not @0EA
# stale the Legacy (MVS conventional data) directory cache if an @0EA
# addition is made to the directory. This causes the next @0EA
# READDIR operation to access the directory @0EA
# information from the Physical File System (PFS) rather than the @0EA
# directory cache. @0EA
# @0EA
# The NORDRCACHE attribute does not apply to z/OS UNIX files. @0EA
# @0EA
# NOTE: When NORDRCACHE is not specified, the addition of an entry @0EA
# to the LEGACY directory cache will not be visible to client @0EA
# until the next readdir cache timeout or a remove from that @0EA
# directory. @0EA
# When NORDRCACHE is specified, the addition will be visible to @0EA
# the client by the subsequent READDIR whether the readdir @0EA
# cache timeout has expired or not. This may impact @0EA
# performance because the directory list must be read from the @0EA
# Physical File System after any addition to the directory. @0EA
# @OEA
# The syntax is: @OEA
# @OEA
# NORDRCACHE # @OEA
# The RDRVERF attribute tells the server to do cookie verifier @0EA
# checking for NFS version 3 readdir and readdirplus requests, @0EA
# and for NFS version 4 readdir requests. @0EA
# NOTE: RDRVERF provides consistency in the listing of a @010A
# directory's content and as a result may impact performance. @010A
# @0EA
# The NORDRVERF attribute tells the server not to do cookie @0EA
# verifier checking for NFS version 3 readdir and readdirplus @0EA
# requests, and for NFS version 4 readdir requests. @0EA
# NOTE: NORDRVERF does not provide consistency in the listing of @010A
# a directory's content and may cause duplicate or omit entries @010A
# when the directory is changing during the listing. @010A
# The default is NORDRVERF. @0EA
nordrverf # @0EA
# ZNFSCLIENT specifies that the NFS Client is z/OS. @LAHA
# @LAHA
# This value is used for compatibility purposes only. For @LAHA
# example, z/OS Unix System Services has separate File and @LAHA
# Directory Default Access Control Lists (ACLs). Other platforms @LAHA
# a common Default ACL which is used for both. This flag @LAHA
# allows the z/OS NFS Server to customize its response to the @LAHA
# NFS Client accordingly. @LAHA
# @LAHA
# NOTE: This value should never be specified by the customer. @LAHA
# this attribute is automatically appended to the procssing @LAHA
# attribute string by the z/OS NFS Client when it detects that @LAHA
# it is sending the mount request (or the last LOOKUP for an @LAHA
# NFS version 4 mount) to a z/OS NFS Server. @LAHA
# If this attribute is specified on a non-z/OS Client Mount @LAHA
# request, some requests may not function properly. @LAHA
# @LAHA
# ZNFSCLIENT DO NOT SPECIFY! The attribute is listed here for @LAHA
# documentation purposes only. @LAHA
#######################################################################
# The following are known as site attributes. #
#######################################################################
# The following are attributes specifiable ONLY in the site
# file (this file).
#
# Some of these values control internal structures and processing
# within the NFS server. Tuning of these values to improve performance
# should be done incrementally and tested.
# SECURITY attribute control the level of security checking.
# The format of the security keyword is security(mvs,hfs,public) @L5DA
# where: @L5DA
# mvs - security option for mvs data access @L5DA
# hfs - security option for HFS data access @L5DA
# public - security option for data access with the public @L5DA
# filehandle @L5DA
# The first positional parameter is required and the other two @L5DA
# are optional. When the optional parameters are not specified @L5DA
# they are assigned the same security as the first parameter. @L5DA
# Four options can be chosen from. They are:
# NONE - No security checking is performed.
# SAF - SAF checking is performed in line.
# EXPORTS - EXPORTS file is used to check security.
# SAFEXP - Both SAF and EXPORTS file checks are performed.
#
#
# Defaults are SAFEXP for all data accesses. @L5DA
security(safexp,safexp,safexp) # @L5DC
# PCNFSD tells the server to start PCNFS server.
# NOPCNFSD tells the server not to start PCNFS server.
# If not specified PCNFSD, default is NOPCNFSD.
nopcnfsd
# LEADSWITCH tells the server to return '/' as the first character
# in each export entry.
# NOLEADSWITCH tells the server not to return '/' as the first character
# in each export entry.
# If not specified NOLEADSWITCH, default is LEADSWITCH.
leadswitch
# MINTIMEOUT and MAXTIMEOUT set allowable values for
# ATTRTIMEOUT, READTIMEOUT, and WRITETIMEOUT. Specify NOMAXTIMEOUT
# to allow NOATTRTIMEOUT, NOREADTIMEOUT and NOWRITETIMEOUT specification
# by clients.
#
# Valid range is from 1 to 32767.
mintimeout(1)
nomaxtimeout
# The logout time is the time limit on inactivity for a given user
# on a machine. When the limit is reached, the user is automatically
# logged out. The user must then do another "mvslogin" to restart
# the session. The time value is specified in seconds.
#
# You would probably want to set LOGOUT to the TSO timeout value that
# is defined at your site.
#
# MAXTIMEOUT does not affect the LOGOUT site attribute.
#
logout(1800) # 30 minutes (30 * 60)
# The readdirtimeout is a new timeout attribute to control the @03A
# timeout of the readdir cache used by MVS conventional data @03A
# sets. The timeout value controls how long before the readdir @03A
# results saved in cache are discarded. @03A
# @03A
# Valid range is from 1 to 32,767 seconds. @04C
# @03A
# n can go as low as 1 second but to avoid the possibility of @04C
# client hanging (because of network delays and staled cache), @04C
# n is not recommended to be lower than 5 seconds. @04A
# n may need to be increased if the network is slow and the @03A
# accessed directory has a lot of entries. @03A
# The default readdirtimeout is 30 seconds. @03A
readdirtimeout(30) # @03A
# The NFSTASKS(n,m,o,t,u) @P0DC
# defines the number of NFS tasks (or threads) to spawn. @P0DC
# @LA1A
# If NFSTASKS(n,m) is specified, then the following is true: @LA1A
# @LA1A
# 'n' is the number of subtasks which handle the asynchronous I/O @L3BC
# operations or short blocking operations (the maximum number of @LA1C
# concurrent NFSS requests) in the z/OS MVS data path. @P0IC
# 'm' is the number of subtasks which handle the long blocking @L3BC
# operations (the maximum number of concurrent NFSS recall and @L3BC
# HFS requests.). Increase this value if your server supports @L3BC
# many active recall operations or HFS clients. @P0IC
#
# Valid range for 'n' is from 4 to 99. @P0IC
# Valid range for 'm' is from 4 to 100. @P0IC
# The sum of 'n' plus 'm' must be less than or equal to 100. @0DC
# @0DA
# NOTE: Based on system resources available below the 16Mb line, @0DA
# the maximum 'n' value may not be achievable. The @0DA
# precise maximum value will be system configuration @0DA
# dependent. If an 878 or 80A Abend is experienced during @P0AC
# NFSS startup, use a smaller value for 'n'. @P0AC
#
# If NFSTASKS(n,m,o) or NFSTASKS(n,m,o,t,u) is specified, @P0DC
# then the following is true: @P0DA
# @LA1A
# 'n' is the number of subtasks which handle the asynchronous I/O @LA1A
# operations or short blocking operations (the maximum number of @LA1A
# concurrent NFSS requests) in the MVS data path. @P0IC
# 'm' is the number of subtasks which handle HFS requests. @LA1A
# Increase this value if your server supports lots of active @LA1A
# HFS clients. @LA1A
# 'o' is the number of subtasks which handle the long blocking @LA1A
# operations (the maximum number of concurrent NFSS recall @LA1A
# requests) in the z/OS MVS data path. @P0GC
# Increase this value if your server supports many active recall @P0IC
# operations. @P0IC
# 't' is the number of transport subtasks which handle TCP @P0DA
# network requests. @P0DA
# 'u' is the number of transport subtasks which handle UDP @P0DA
# network requests. @P0DA
#
# Valid range for 'n' is from 4 to 99. @P0IC
# Valid range for 'm' is from 4 to 100. @P0IC
# Valid range for 'o' is from 1 to 99. @0DC
# Valid range for 't' is from 4 to 32. @P0DA
# Valid range for 'u' is from 4 to 32. @P0DA
# The sum of 'n' plus 'o' must be less than or equal to 100. @0DC
# @0DA
# NOTE: Based on system resources available below the 16Mb line, @0DA
# the maximum 'n' + 'o' value may not be achievable. The @0DA
# precise maximum value will be system configuration @0DA
# dependent. If an 878 or 80A Abend is experienced during @P0AC
# NFSS startup, use a smaller value for 'n' + 'o'. @P0AC
nfstasks(8,16,8,4,4) # @P0DC
# @L66A
# The MINTASKS(n,m,o) defines the minimum number of NFS tasks @L66A
# or threads) allowed to run. Tasks may be terminated for reasons @L66A
# such as 878 or 80A abends. @L66A
#
# Note: Options t and u in NFSTASKS(n,m,o,t,u) do not have @P0DA
# MINTASKS counterparts because MINTASKS only applies to @P0DA
# the request processing tasks, not the transport tasks. @P0DA
#
# 'n' is the minimum number of subtasks which handle @L66A
# the asynchronous I/O operations or short blocking @L66A
# operations. If the number of active 'short' tasks becomes @L66A
# less than 'n' the shutdown process of NFS server will start.@L66A
# 'm' is the minimum number of subtasks which handle HFS @L66A
# requests. If the number of active HFS tasks becomes less @L66A
# than 'm' the shutdown process of NFS server will start. @L66A
# 'o' is the minimum number of subtasks which handle long @L66A
# blocking operations. If the number of active legacy long @L66A
# service tasks becomes less than 'o' the shutdown process @L66A
# of NFS server will start. @L66A
# If 'n','m' or 'o' are greater than the corresponding values @L66A
# in NFSTASKS, they are assigned to half of the NFSTASKS value. @P0IC
# If 'n','m' or 'o' are not specified they are assigned default @P0IC
# values of 4,4,1 respectively. @P0IC
# Valid range for 'n' is from 4 to 99. @P0IC
# Valid range for 'm' is from 4 to 100. @P0IC
# Valid range for 'o' is from 1 to 99. @0DC
# The sum of 'n' plus 'o' must be less than or equal to 100. @0DC
# @0DA
# NOTE: Based on system resources available below the 16Mb line, @0DA
# the maximum 'n' + 'o' value may not be achievable. The @0DA
# precise maximum value will be system configuration @0DA
# dependent. If an 878 or 80A Abend is experienced during @P0AC
# NFSS startup, use a smaller value for 'n' + 'o'. @P0AC
mintasks(4,8,4) # @L66A
# The RESTIMEOUT(n,m) defines resource cleanup timer. @L3LA
# 'n' is the resource retention period for mounts and
# associated resources which will be removed if the mounts
# have been inactive more than 'n' number of hours. @L3LA
# 'm' is the time of day to do the cleanup work
# for mounts and associated resources which
# have been inactive more than 'n' number of hours.
# The time of day is specified as a 24 hour local time value
# The starting time is at least 24 hours from NFSS start-up.
# @L3LA
# NFSS will appear slow during this cleanup activity. @L3LA
# Cleanup will be executed under main task, thus preventing @L3LA
# any additional work from executing. @L3LA
# The value of RESTIMEOUT should be set to a value @L3LA
# such that this cleanup activity will occur while @L3LA
# NFSS is lightly loaded. @L3LA
# @L3LA
# Valid range for 'n' is from 1 to 720. @L3LA
# If n is set to 0, NFSS will not remove any mount points. @L3LA
# Note: NFSS keeps all the information for the @L3LA
# inactive mount points, thus creating long chains @L3LA
# to be searched. @L3LA
# @L3LA
# Valid range for 'm' is from 0 to 23. @L3LA
restimeout(48,0) # @L3LA
# The CACHEWINDOW attribute limits the number of windows to be used
# for each data set in caching client block writes which are received
# out of order.
# The size of each window is determined by the packet size.
# The suggested value is some small multiple of
# the number of BIODs running on an NFS client.
#
# If the "adjusted" LOGICALCACHE is larger than 2GB then the @LEPA
# specified CACHEWINDOW value is ignored. @LEPA
# Valid range is from 1 to 2048. @011C
#
cachewindow(112) # @P5C
# The HFSPREFIX (or HFS) attribute specifies a z/OS UNIX file @LBBC
# system prefix to be imbedded in the mount directory path name. @LBBC
# The default value of the z/OS UNIX file system prefix is /hfs. @LBBC
# Mount requests received by the Network File System beginning @LBBC
# with this prefix value are identified as mount requests for @LBBC
# z/OS UNIX. The prefix value is not part of the path name. @LBBC
# @01A
# Note: The file system must be mounted locally by z/OS UNIX. @LBBC
# Otherwise, the client mount fails. @LBBC
# hfsprefix(prefix) or hfs(prefix) @LBBC
hfsprefix(/hfs) # @LBBC
# The MVSPREFIX attribute specifies an MVS data set prefix @LBBA
# to be imbedded in the mount directory path name. @LBBA
# The default value of the MVS data set prefix is /mvs. @LBBA
# Mount requests received by the Network File System beginning @LBBA
# with this prefix value are identified as mount requests for @LBBA
# MVS data sets. The prefix value is not part of the path name. @LBBA
# @LBBA
# mvsprefix(prefix) @LBBA
mvsprefix(/mvs) # @LBBA
# The IMPPREFIX attribute specifies how mount directory path @LBBA
# names which do not have either an HFSPREFIX or an MVSPREFIX @LBBA
# are to be interpreted. If no prefix is present in the path @LBBA
# name, the path is to be interpreted according to the IMPPREFIX @LBBA
# setting: @LBBA
# HFS - assume it is a z/OS UNIX file system. @LBBA
# MVS - assume it is an MVS data set. @LBBA
# HFS,MVS - try z/OS UNIX file system first. If that fails @LBBA
# or is not found, try MVS data set. @LBBA
# MVS,HFS - try MVS data set first. If that fails or is not @LBBA
# found, try z/OS UNIX file system. @LBBA
# The default value is MVS, which matches pre-z/OS V1R11 @LBBA
# functionality. @LBBA
# @LBBA
# Note: For the MVS,HFS setting, the MVS data set must exist @LBBA
# if the path is referring to an MVS data set. Prior to @LBBA
# this attribute being added, it was not necessary for the @LBBA
# data set to pre-exist. However, now, if it does not @LBBA
# exist, the 2nd option (HFS) will be used. @LBBA
impprefix(mvs) # @LBBA
# fileidsize site attribute: @P08C
# @P08C
# The fileidsize attribute tells NFS how to handle the size @P08C
# of fileids. Fileids can be recognized either as 32-bit @P08C
# values or 64-bit values. @P08C
# @P08C
# Client platforms often copy the returned NFS object fileid to @P08C
# client internal structure fields such as a UNIX inode. Older @P08C
# 32-bit applications and operating systems may not support NFS @P08C
# fileids larger than 32 bits resulting in value too large errors @P08C
# or unexpected client behaviors. @P08C
# @P08C
# IBM recommends upgrading older applications and operating @P08C
# systems to an environment which supports the latest supported @P08C
# NFS Protocol Version offering the largest supported fileidsize. @P08C
# If upgrading your environment is not possible, the fileidsize @P08C
# may be set to a lower value to support older applications if @P08C
# accessed data can be properly recovered or refreshed. @P08C
# @P08C
# Valid values are 32 or 64. @P08C
# @P08C
# NFS Version 2 maximum value in effect is (32) bits @P08C
# NFS Version 3 maximum value in effect is (32) bits @P08C
# NFS Version 4 maximum value in effect is (64) bits @P08C
# @P08C
# If NFS Version 2 or Version 3 is in effect, the maximum value @P08C
# of 32 is used. When the specified fileidsize value @P08C
# is larger than the maximum value for the NFS Version in effect, @P08C
# the proper maximum value will be used for generating @P08C
# the NFS fileid attribute. @P08C
# @P08C
# Warning: @P08C
# @P08C
# The fileidsize site attribute should be set to the maximum @P08C
# value to generate the maximum size in unique ids for NFS @P08C
# objects within a file system. Setting the value lower than @P08C
# the allowed maximum (e.g. NFS Version 4) may result in the @P08C
# same NFS fileid to be returned to the client for @P08C
# different server objects. As a result, reducing the @P08C
# fileidsize site attribute for client @P08C
# compatibility may result in data corruption. @P08C
# @P08C
# The default is fileidsize(64) @P08C
fileidsize(64) # @P08C
# The LOGICALCACHE attribute sets the high water mark for all the
# cache windows combined(in bytes).
# The LOGICALCACHE attribute denotes the memory above 2GB BAR @LEPA
# and it is related to the system MEMLIMIT. @LEPA
# *) The z/OS Server will terminate if the specified LOGICALCACHE @LEPA
# is LARGER THAN the usable memory of the system MEMLIMIT. @LEPA
# *) The z/OS Server may adjust the LOGICALCACHE to take @LEPA
# advantage of the large usable memory of the system MEMLIMIT @LEPA
# ¦ = MIN( 4096G, 1/2 of available memory above 2GB ) ¨. @LEPA
# *) The modify operator BUFFERUSAGE command reports the @LEPA
# "adjusted" LOGICALCACHE. @LEPA
# @LEPA
# Valid range is from 1 to 4096G. @LEPC
# The default value is 16MB @P0LA
logicalcache(4G) # @LEPC
# BUFHIGH(XX,YY) @LCQC
# The BUFHIGH attribute sets the high water mark (in bytes) @LCQC
# of data buffers ( below 2GB BAR ) @LEPC
# before a buffer reclamation takes place @LEPC
# (it's the first parameter XX) and the watermark in percent of @LCQC
# XX for printing a data buffer utilization alert message @LCQC
# (it's the second parameter YY). @LCQC
# A higher number of XX means more caching and probably better @LCQC
# read performance. @LCQC
# (If XX is larger than the available storage @LEPC
# in the extended private area (implied by the REGION parameter @LCQC
# coded in your PROC) at startup, the NFS server will @LCQC
# shutdown immediately.) @LCQC
#
# At z/OS Server startup, the actual value (xx) specified with @P0GA
# bufhigh may be adjusted by the z/OS NFS Server internally @P0GA
# depending on the available region size and other z/OS NFS @P0GA
# Server memory requirements to enable the z/OS NFS Server to @P0GA
# execute properly. @P0GA
#
# The valid value of YY is from 50 to 90 percents or 0(zero) @LCQC
# to turn off the data buffer utilization reporting mechanism. @LCQC
# Any invalid value of YY is converted to 80 percent. @LCQC
# The default value of YY is 80 percent. @LCQC
# If Bufhigh contains only the first parameter (comma and YY are @LCQC
# ommited) the default value of YY is used. @LCQC
bufhigh(32M,80) # @P0IC
# On reaching the BUFHIGH high water mark, a percentage of buffers
# is reclaimed for reuse. This is the PERCENTSTEAL. A higher value
# means a reclaim operation is frequently performed, and that
# the cached buffers will be significantly trimmed on each reclaim.
# This can result in a poor read performance, because readahead
# buffers may be stolen.
# Lower values result in less frequent reclaim operations, and the
# cached buffers are not significantly trimmed on each reclaim.
#
# Valid range is from 1 to 99.
percentsteal(20)
# The READAHEADMAX value defines the maximum read ahead (in bytes)
# during read processing.
# This reduces the amount of synchronous physical I/O required for
# NFS read requests in sequential processing. It also reduces context
# switching overhead on NFS read requests by allowing more read
# requests to be satisfied directly from the main task.
#
# The number is usually set to 2 to 4 times the common block size
# used for file access (which is recommended at 8K for AIX file
# activity).
#
# Valid range is from 1 to 128K
readaheadmax(16K)
# The MAXRDFORSZLEFT attribute defines the number of physical block
# buffers to be remained after a read-for-size operation.
# These buffers are remained to satisfy potential subsequent
# NFS read requests against the same file.
# The buffers remained are subject to trimming during buffer
# steal operations.
#
# Valid range is from 1 to 1024.
maxrdforszleft(32)
# SMF(level,switch) attribute controls the level of smf support @08A
# and defines whether to start SMF records collection @08A
# at the NFSS startup or not. @08A
# The following 'level' options are available: @LCOC
# NONE - No smf records are to be produced. @08A
# ALL - All smf records are to be produced. @LCOA
# USERFILE - Both user session and file usage smf records are @LCOM
# to be produced. @LCOM
# List of levels that are delimited by commas, can be specified. @LCOA
# At least one of the levels: ( USER | FILE | AUDIT ) @LCOA
# must be specified for a list, and the remaining levels @LCOA
# are optional. Where: @LCOA
# USER - User session smf records are to be produced. @08A
# FILE - File usage smf records are to be produced. @08A
# AUDIT- File creation, file removing and file renaming @LCOA
# smf records are to be produced. @LCOA
# Default level is NONE. @LCOC
# Switch has two options: @08A
# ON - activates SMF records collection at the NFSS @08A
# startup @08A
# OFF - activation of SMF records collection can be done @08A
# manually by issuing the operator command MODIFY. @08A
# Switch is optional parameter. Its default value is OFF. @08A
# @08A
# The syntax of the SMF attribute is @08A
# SMF(NONE|ALL|USERFILE|subtype_list,ON|OFF) @LCOC
# where subtype_list is list of 1 - 3 levels:(USER|FILE|AUDIT) @LCOA
# separeted by commas. @LCOA
# For example, @08A
# SMF(USER,ON) @08A
# SMF(AUDIT,FILE,OFF) @LCOA
# You can use the short form of the SMF attribute: @08A
# SMF(level) @08A
# In that case the value of the switch is OFF @08A
smf(none,off) # @08A
# SFMAX(n) where n specifies the maximum size (in kilobytes) of @L59A
# allocated storage to hold all of the side files. n is an @L59A
# integer from 0 to 2000 (2MB). The default value is 0 and it @L59A
# also signifies that no mapping is allowed on this NFS server. @L59A
# If SFMAX=0, the specification of sidefile in the attributes @L59A
# data set will cause the server to shut down and the @L59A
# specification of sidefile in any subsequent MOUNT commands will @L59A
# cause the mount to fail as mapping is not allowed on this NFS @L59A
# server. If the amount of storage specified can not be obtained @L59A
# during server initialization then the server will shut down @L59A
# immediately. @L59A
# @L59A
# The default is SFMAX(0). @L59A
sfmax(0) # @L59A
# PUBLIC(legacy_path,hfs_path) specifies the legacy path @L53A
# (MVS conventional data) and/or HFS path that is associated with @L53A
# the public file handle for WebNFS access. The first path, if @L53A
# specified, is the legacy path. The second path is the HFS @L53A
# path. MVS and HFS prefix values are optional on these path @LBBC
# name values since they are positional and thus self-defining. @LBBA
# However, if a prefix is specified, it must match the active @LBBA
# prefix value to be recognized as such. Otherwise, it will @LBBA
# be assumed to be a qualifier in the path name. The PUBLIC @LBBA
# keyword must be specified after the HFSPREFIX (or HFS) and @LBBM
# MVSPREFIX keywords in this site attribute table. @LBBM
# If the first path is not there, a comma must precede @LBBC
# the second path. If the PUBLIC keyword is specified then one @L53A
# of the paths must be specified. @LBBC
# A LOOKUP request with the public file handle will determine @L53A
# which of the two paths it is referring to by the pathname that @L53A
# it comes in with. An absolute pathname will tell the server @L53A
# which of the paths it is referring to by a match on one of the @L53A
# paths specified. A LOOKUP request with a relative pathname @L53A
# will be taken to be an HFS request if HFS is active (i.e. an @L53A
# hfs_path has been provided); otherwise, it is treated as a @L53A
# legacy request. @L53A
# @L53A
# The default is no PUBLIC paths. @L53A
# CHECKLIST refers to the data set pointed to by the EXPORTS DD @L81C
# in the startup procedure to be scanned. The data set may contain@L81C
# a list of CHKLIST entries which should match a subsequent @L81C
# mount point or be the parent of a subsequent mount point to @L5DA
# allow SAF checking to be bypassed for everything underneath @L5DA
# that mount point. CHECKLIST is only valid if SAF checking is @L5DA
# the security option for the particular data access; otherwise, @L5DA
# it is ignored even if specified. See GFSAPEXP in NFSSAMP @L81C
# library for sample CHKLIST entries description. @L81C
# NOCHECKLIST will cause the server not to look at the CHKLIST @L81C
# information in the EXPORTS data set even if it is there. This @L81C
# is the default. @L81C
nochecklist # @L5DA
# FN_DELIMITER specifies a character to be used instead of a @L5XA
# comma to delimit the file name from the accompanying attributes.@L5XA
# This will allow those sites which have UNIX data sets containing@L5XA
# commas to copy and store their data on the Z/OS NFS Server. @L6C
# @L5XA
# An example of this would be: @L5XA
# The attribute data set contains fn_delimiter(;) @L5XA
# this changes the default delimiter character from a comma @L5XA
# to a semicolon. @L5XA
# @L5XA
# so instead entering ... vi "data_set_name,text,lf" @L5XA
# @L5XA
# you would enter.........vi "comma,in-name;text,lf" @L5XA
# @L5XA
# This is a new SITE attribute and it's default is a comma @L5XA
# @L5XA
fn_delimiter(,) # @L5XA
# The hfsfbtimeout is a new timeout attribute to control the @06A
# timeout of the HFS vnode token used by NFS server. @06A
# The timeout value controls how long before vnode tokens @06A
# saved in file blocks are released. @06A
# @06A
# Valid range is from 1 to 32,767 seconds. @06A
# @06A
# n can go as low as 1 second but to avoid the possibility of @06A
# client hanging (because of network delays), @06A
# n is not recommended to be lower than 5 seconds. @06A
# n may need to be increased if the network is slow and the @06A
# accessed directory has a lot of entries. @06A
# The default hfsfbtimeout is 60 seconds. @06A
# @06A
hfsfbtimeout(60) # @06A
# The UPCASE attribute will convert the messages to uppercase. @07A
# The MIXCASE attribute will display the messages without @07A
# any modification (as they are). @07A
# The default is upcase. @07A
# @07A
upcase # @07A
# NOREC878 specifies that recovery processing of 878 or 80A @L66A
# abends will be turned OFF. That is, if this type of abend @L66A
# occurs, server will shutdown without recovery. @L66A
# It can be used for Debug only. @L66A
# REC878 specifies that recovery processing of 878 or 80A @L66A
# abends will be turned ON, and affected tasks will attempt to @L66A
# recover. @L66A
# This is default value. @L66A
rec878 # @L66A
# NLM tells the server to start Lockd/Statd. @L74A
# NONLM tells the server not to start Lockd/Statd. @L74A
# If not specified NLM, default is NONLM. @L74A
nonlm # @L74A
# DELEGATION tells the server to enable delegation support. @LBCA
# NODELEGATION tells the server not to enable delegation support. @LBCA
# If neither is specified, the default is NODELEGATION. @LBCA
# nodelegation # @LBCA
# Stringprep Support:
# Stringprep is a standard by which similar characters are @L82A
# mapped to a common character to improve string compares. NFS @L82A
# calls Z/OS Coversion Services stringprep services when @L82A
# receiving network strings such as filenames. @L82A
# STRINGPREP tells the server to invoke Stringprep support. @L82A
# NOSTRINGPREP tells the server to disable Stringprep support. @L82A
# Default: @L82A
# For compatibility with NFS version 4 support in Z/OS V1R7 @L82A
# stringprep is disabled by default: NOSTRINGPREP @L82A
# @L82A
# stringprep # @L82A
# nostringprep # @L82A
nostringprep # @L82A
# Leasetime(n) specifies the lease time for NFSv4 access. The @L74A
# lease time is the length of time for which a client is @L74A
# irrevocably granted a lock. At the end of a lease period the @L74A
# lock may be revoked if the lease has not been extended. @L74A
# The lock must be revoked if a conflicting lock has been @L74A
# granted after the lease interval. The lease time is also @L74A
# used to specify the grace period duration. The grace period @L74A
# is the length of time provided for NFSv4 and NLM clients to @L74A
# reclaim locks following a server crash and restart. @L74A
# @L74A
# n is from 5 seconds to 300 seconds. @L74A
# The default value is 120 seconds. @L74A
# n must be smaller than logout value if logout is not 0. @L74A
# @L74A
# NOTE: When using the NFS version 4 protocol, the lease time @P09A
# value should be set to a value larger than or equal to @P09A
# the attr/read/write timeout values. Otherwise, it is @P09A
# possible for performance problems to occur when @P09A
# attempting to access MVS data sets. @P09A
leasetime(120) # @L74A
# The REMOUNT attribute enables the NFS Server to process NFS @0AA
# requests after the NFS Server is restarted even though the HFS @0AA
# file system was remounted with a new HFS file system number @0AA
# (device number) after its last usage. @0AA
# Use of the REMOUNT attribute causes the NFS Server to @0AA
# automatically access a remounted HFS file system even though it @0AA
# may have been changed prior to remounting. @0AA
# Any active client mounts are reestablished. @0AA
# The NOREMOUNT attribute enables the NFS Server to fail NFS @0AA
# requests (with return value NFSERR_STALE) if the HFS file system @0AA
# was remounted with a new HFS file system number (device number) @0AA
# after its last usage. @0AA
# The default is NOREMOUNT. @0AA
# This site attribute applies only to HFS file systems. @0AA
# @0AA
noremount # @0AA
# DHCP specifies that NFS Clients using Dynamic IP addresses @L76A
# are supported. To function properly, this support requies @L76A
# that the Clients have static Host Names and that they @L76A
# dynamically update the DNS server with their IP address @L76A
# changes. @L76A
# NODHCP specifies that only NFS Clients using Static IP @L76A
# addresses are supported. @L76A
# The default is NODHCP. @L76A
# @L76A
nodhcp # @L76A
# DENYRW specifies that the NFS Version 4 OPEN operation, @L74M5A
# and the NLM Protocol NLM_Share procedure, should support @L74M5A
# the Windows Share_Deny value. @L74M5A
# NODENYRW specifies that the NFS Version 4 OPEN operation, @L74M5A
# and the NLM Protocol NLM_Share procedure, should ignore @L74M5A
# the Windows Share_Deny value for z/OS UNIX file systems (the @L7EC
# value will be treated as if DENY_NONE was specified). @L74M5A
# The default is DENYRW. @L74M5A
# @P07C
DENYRW # @P07C
# mvssec( , , , ) specifies the permissible network @L74AA
# transmission security for MVS data. @L74AA
# The valid flavors are krb5,krb5i,krb5p,sys. All or none of @L74AA
# them can be specified. @L74AA
# The default value is to allow all pseudo flavors for MVS @L74AA
# Data sets. @L74AA
# If mvssec is specified, then only the specified flavors @L74AA
# will be allowed for MVS Data sets. @L74AA
# @L74AA
mvssec(krb5,krb5i,krb5p,sys) # @L74AA
# hfssec( , , , ) specifies the permissible network @L74AA
# transmission security for HFS data. @L74AA
# The valid flavors are krb5,krb5i,krb5p,sys. All or none of @L74AA
# them can be specified. @L74AA
# The default value is to allow all pseudo flavors for MVS @L74AA
# Data sets. @L74AA
# If hfssec is specified, then only the specified flavors @L74AA
# will be allowed for HFS Data sets. @L74AA
# @L74AA
hfssec(krb5,krb5i,krb5p,sys) # @L74AA
# pubsec( , , , ) specifies the permissible network @L74AA
# transmission security for PUBLIC data. @L74AA
# The valid flavors are krb5,krb5i,krb5p,sys. All or none of @L74AA
# them can be specified. @L74AA
# The default value is to allow all pseudo flavors for MVS @L74AA
# Data sets. @L74AA
# If pubsec is specified, then only the specified flavors @L74AA
# will be allowed for PUBLIC Data sets. @L74AA
# @L74AA
pubsec(krb5,krb5i,krb5p,sys) # @L74AA
# NFSV4DOMAIN(NFSv4_default_domain) specifies the "pseudo" NFSv4 @015A
# domain for the NFS v4 protocol (NFSv4) name mapping. The @015A
# "pseudo" NFSv4 domain allows various NFSv4 Clients from various @015A
# network domain to seamlessly access the server provided that @015A
# these NFSv4 Clients are also configured with the same @015A
# "NFSv4_default_domain". @015A
# When the NFSV4DOMAIN is omitted, the participating NFSv4 @015A
# Clients' domain must match the Server's network domain for @015A
# the proper NFS V4 protocol name mapping. @015A
# NFSV4DOMAIN(NFSv4_default_domain) # @LAGA
#DlyDTimeout(10) Specifies the minimum Delay Detection time value @LBKA
# in sec before the Delay Detection mechanism observes a delay in @LBKA
# an external call/function and prints message GFSA1030W on @LBKA
# console. @LBKA
# The maximum value for DlyDTimeout is 60 seconds. @LBKA
# The minimum value for DlyDTimeout is 5 seconds. @LBKA
# Any value of DlyDtimeout from 1 to 4 seconds is @P0FC
# converted to 5 seconds. @LBKA
# The Default value is 10 seconds. @LBKA
# If DlyDTimeout is set to 0 the DelayDetection mechanism @LBKA
# is turned OFF @LBKA
# DlyDTimeout(10) # @LBKA
# setgid(POSIX | ZOSUNIX) attribute controls GID inheritance for @012A
# new UNIX objects. @012A
# POSIX - z/OS NFS server @012A
# uses POSIX rules in GID inheritance for new z/OS UNIX objects. @012A
# If the S_ISGID bit of the parent directory is on, @012A
# then the new GID is set to the GID of the parent directory. @012A
# Otherwise, it is set from the GID of the process. @012A
# A new directory inherits the S_ISGID bit value @012A
# of the parent directory. @012A
# ZOSUNIX - z/OS NFS server @012A
# provides the compatibility with z/OS UNIX. @012A
# When the RACF profile FILE.GROUPOWNER.SETGID in the UNIXPRIV @012A
# class is set, z/OS NFS server @012A
# uses POSIX rules, as stated above. Otherwise, a @012A
# new GID is always set to the GID of the parent directory, @012A
# and for a new directory, the S_ISGID bit is always set off. @012A
# NOTE: Some NFS clients (e.g. SUN, AIX) force GID setting after @012A
# object creation and prevent compatibility with @012A
# z/OS UNIX even though the setgid(ZOSUNIX) attribute is set. @012A
# @012A
# The default is POSIX. @012A
setgid(POSIX) # @012A
# SYMRESOLVE specifies that the server will resolve a symlink @018A
# found in EXPORT or CHECKLIST entries. @018A
# The new EXPORT or CHECKLIST entries are created in memory only @018A
# This option is only for NFSv4 LOOKUP in HFS space when @018A
# a symlink is found within an export entry. @018A
# The default is NOSYMRESOLVE. @018A
# nosymresolve # @018A
#mvslogindelay() Specifies the delay time value in seconds since @019A
# z/OS NFS Server startup. To avoid an NFS client's cache @019A
# invalidation due to access errors after z/OS NFS server startup,@019A
# in SECURITY(SAF/SAFEXP) mode, the server maps the reply error @019A
# NFSERR_ACCES to NFSERR_DELAY on NFS v3 or NFS v4 RPCs, until @019A
# the mvslogin is received or the mvslogindelay expires. @019A
# This allows an NFS client time to reissue an mvslogin. @019A
# The maximum value for mvslogindelay is 300 seconds. @019A
# The minimum value for mvslogindelay is 0 seconds. @019A
# The Default value is 0 seconds (off). @019A
# If mvslogindelay is set to 0 the error mapping is turned OFF. @019A
#mvslogindelay(0) # @019A
|
z/OS Network File System Guide and Reference
Copyright IBM Corporation 1990, 2014