shaba/lvm2
1
0
Fork 0
mirror of git://sourceware.org/git/lvm2.git synced 2025-01-04 09:18:36 +03:00
Code Releases Activity
dev-dct-cmd-defs28
lvm2/scripts/command-lines.in
David Teigland 8e85c41dc7 commands: new method for defining commands 
. Define a prototype for every lvm command.
. Match every user command with one definition.
. Generate help text and man pages from them.

The new file command-lines.in defines a prototype for every
unique lvm command.  A unique lvm command is a unique
combination of: command name + required option args +
required positional args.  Each of these prototypes also
includes the optional option args and optional positional
args that the command will accept, a description, and a
unique string ID for the definition.  Any valid command
will match one of the prototypes.

Here's an example of the lvresize command definitions from
command-lines.in, there are three unique lvresize commands:

lvresize --size SizeMB LV
OO: --alloc Alloc, --autobackup Bool, --force,
--nofsck, --nosync, --noudevsync, --reportformat String, --resizefs,
--stripes Number, --stripesize SizeKB, --test, --poolmetadatasize SizeMB
OP: PV ...
ID: lvresize_by_size
DESC: Resize an LV by a specified size.

lvresize LV PV ...
OO: --alloc Alloc, --autobackup Bool, --force,
--nofsck, --nosync, --noudevsync,
--reportformat String, --resizefs, --stripes Number, --stripesize SizeKB,
--test
ID: lvresize_by_pv
DESC: Resize an LV by a specified PV.

lvresize --poolmetadatasize SizeMB LV_thinpool
OO: --alloc Alloc, --autobackup Bool, --force,
--nofsck, --nosync, --noudevsync,
--reportformat String, --stripes Number, --stripesize SizeKB,
--test
OP: PV ...
ID: lvresize_pool_metadata_by_size
DESC: Resize a pool metadata SubLV by a specified size.

The three commands have separate definitions because they have
different required parameters.  Required parameters are specified
on the first line of the definition.  Optional options are
listed after OO, and optional positional args are listed after OP.

This data is used to generate corresponding command definition
structures for lvm in command-lines.h.  usage/help output is also
auto generated, so it is always in sync with the definitions.

Example of the corresponding generated structure in
command-lines.h for the first lvresize prototype
(these structures are never edited directly):

commands[85].name = "lvresize";
commands[85].command_line_id = "lvresize_by_size";
commands[85].command_line_enum = lvresize_by_size_CMD;
commands[85].fn = lvresize;
commands[85].ro_count = 1;
commands[85].rp_count = 1;
commands[85].oo_count = 22;
commands[85].op_count = 1;
commands[85].desc = "DESC: Resize an LV by a specified size.";
commands[85].usage = "lvresize --size Number[m|unit] LV"
" [ --resizefs, --poolmetadatasize Number[m|unit], COMMON_OPTIONS ]"
" [ PV ... ]";
commands[85].usage_common =
" [ --alloc contiguous|cling|cling_by_tags|normal|anywhere|inherit, --commandprofile String, --config String, --driverloaded y|n, --nosync, --noudevsync, --profile String, --reportformat String, --version, --autobackup y|n, --debug, --force, --help, --stripes Number, --stripesize Number[k|unit], --nofsck, --quiet, --test, --verbose, --yes, --commandprofile String, --config String, --debug, --driverloaded y|n, --help, --profile String, --quiet, --verbose, --version, --yes, --force, --test, --noudevsync ]";
commands[85].required_opt_args[0].opt = size_ARG;
commands[85].required_opt_args[0].def.val_bits = val_enum_to_bit(sizemb_VAL);
commands[85].required_pos_args[0].pos = 1;
commands[85].required_pos_args[0].def.val_bits = val_enum_to_bit(lv_VAL);
commands[85].optional_opt_args[0].opt = commandprofile_ARG;
commands[85].optional_opt_args[0].def.val_bits = val_enum_to_bit(string_VAL);
commands[85].optional_opt_args[1].opt = config_ARG;
commands[85].optional_opt_args[1].def.val_bits = val_enum_to_bit(string_VAL);
commands[85].optional_opt_args[2].opt = debug_ARG;
commands[85].optional_opt_args[3].opt = driverloaded_ARG;
commands[85].optional_opt_args[3].def.val_bits = val_enum_to_bit(bool_VAL);
commands[85].optional_opt_args[4].opt = help_ARG;
commands[85].optional_opt_args[5].opt = profile_ARG;
commands[85].optional_opt_args[5].def.val_bits = val_enum_to_bit(string_VAL);
commands[85].optional_opt_args[6].opt = quiet_ARG;
commands[85].optional_opt_args[7].opt = verbose_ARG;
commands[85].optional_opt_args[8].opt = version_ARG;
commands[85].optional_opt_args[9].opt = yes_ARG;
commands[85].optional_opt_args[10].opt = alloc_ARG;
commands[85].optional_opt_args[10].def.val_bits = val_enum_to_bit(alloc_VAL);
commands[85].optional_opt_args[11].opt = autobackup_ARG;
commands[85].optional_opt_args[11].def.val_bits = val_enum_to_bit(bool_VAL);
commands[85].optional_opt_args[12].opt = force_ARG;
commands[85].optional_opt_args[13].opt = nofsck_ARG;
commands[85].optional_opt_args[14].opt = nosync_ARG;
commands[85].optional_opt_args[15].opt = noudevsync_ARG;
commands[85].optional_opt_args[16].opt = reportformat_ARG;
commands[85].optional_opt_args[16].def.val_bits = val_enum_to_bit(string_VAL);
commands[85].optional_opt_args[17].opt = resizefs_ARG;
commands[85].optional_opt_args[18].opt = stripes_ARG;
commands[85].optional_opt_args[18].def.val_bits = val_enum_to_bit(number_VAL);
commands[85].optional_opt_args[19].opt = stripesize_ARG;
commands[85].optional_opt_args[19].def.val_bits = val_enum_to_bit(sizekb_VAL);
commands[85].optional_opt_args[20].opt = test_ARG;
commands[85].optional_opt_args[21].opt = poolmetadatasize_ARG;
commands[85].optional_opt_args[21].def.val_bits = val_enum_to_bit(sizemb_VAL);
commands[85].optional_pos_args[0].pos = 2;
commands[85].optional_pos_args[0].def.val_bits = val_enum_to_bit(pv_VAL);
commands[85].optional_pos_args[0].def.flags = ARG_DEF_FLAG_MAY_REPEAT;

Every user-entered command is compared against the set of
command structures, and matched with one.  An error is
reported if an entered command does not have the required
parameters for any definition.  The closest match is printed
as a suggestion, and running lvresize --help will display
the usage for each possible lvresize command.

The prototype syntax used for help/man output includes
required --option and positional args on the first line,
and optional --option and positional args enclosed in [ ]
on subsequent lines.

  command_name <required_opt_args> <required_pos_args>
          [ <optional_opt_args> ]
          [ <optional_pos_args> ]

$ lvresize --help
  lvresize - Resize a logical volume

  Resize an LV by a specified size.
  lvresize --size Number[m|unit] LV
        [ --resizefs,
          --poolmetadatasize Number[m|unit],
          COMMON_OPTIONS ]
        [ PV ... ]

  Resize an LV by a specified PV.
  lvresize LV PV ...
        [ --resizefs,
          COMMON_OPTIONS ]

  Resize a pool metadata SubLV by a specified size.
  lvresize --poolmetadatasize Number[m|unit] LV_thinpool
        [ COMMON_OPTIONS ]
        [ PV ... ]

  Common options:
        [ --alloc contiguous|cling|cling_by_tags|normal|anywhere|inherit,
          --commandprofile String,
          --config String,
          --driverloaded y|n,
          --nosync,
          --noudevsync,
          --profile String,
          --reportformat String,
          --version,
          --autobackup y|n,
          --debug,
          --force,
          --help,
          --stripes Number,
          --stripesize Number[k|unit],
          --nofsck,
          --quiet,
          --test,
          --verbose,
          --yes,
          --commandprofile String,
          --config String,
          --debug,
          --driverloaded y|n,
          --help,
          --profile String,
          --quiet,
          --verbose,
          --version,
          --yes,
          --force,
          --test,
          --noudevsync ]

  (Use --help --help for usage notes.)

$ lvresize --poolmetadatasize 4
  Failed to find a matching command definition.
  Closest command usage is:
  lvresize --poolmetadatasize Number[m|unit] LV_thinpool

Man page prototypes are also generated from the same original
command definitions, and are always in sync with the code
and help text.

Very early in command execution, a matching command definition
is found.  lvm then knows the operation being done, and that
the provided args conform to the definition.  This will allow
lots of ad hoc checking/validation to be removed throughout
the code.

Each command definition can also be routed to a specific
function to implement it.  The function is associated with
an enum value for the command definition (generated from
the ID string.)  These per-command-definition implementation
functions have not yet been created, so all commands
currently fall back to the existing per-command-name
implementation functions.

Using per-command-definition functions will allow lots of
code to be removed which tries to figure out what the
command is meant to do.  This is currently based on ad hoc
and complicated option analysis.  When using the new
functions, what the command is doing is already known
from the associated command definition.

So, this first phase validates every user-entered command
against the set of command prototypes, then calls the existing
implementation.  The second phase can associate an implementation
function with each definition, and take further advantage of the
known operation to avoid the complicated option analysis.
2016-10-21 11:44:13 -05:00

1537 lines
48 KiB
Plaintext
Raw Permalink Blame History

#
# When this file is changed, tools/command-lines.h
# and tools/command-lines-count.h must be regenerated
# with:
#
# scripts/create-commands --output count scripts/command-lines.in > tools/command-lines-count.h
# scripts/create-commands --output struct scripts/command-lines.in > tools/command-lines.h
#
#
# Syntax
#
# A new command has a unique combination of:
# command name, required option args and required
# positional args.
#
# To define a new command, begin a single line with a
# command name, followed by required options/args,
# (e.g. --foo, or --foo val), followed by required
# positional args, (e.g. VG)
#
# After the single line of required elements are lines
# of optional elements:
# OO: <optional --option args>
# OP: <optional positional args>
#
# command_name required_opt_arg ... required_pos_arg ...
# OO: optional_opt_arg, ...
# OP: optional_pos_arg ...
#
# required_opt_arg/optional_opt_arg must begin with the
# long form option name, e.g. --foo. If the option name
# takes a value, then the type of value is specified,
# e.g. --foo String.
#
# Possible option names are listed in args.h
#
# Use --foo_long to specify that only the long form of
# --foo is accepted by the command. (This is uncommon.)
#
# Possible option arg types are shown in tools/vals.h,
# e.g. Bool, String, VG, SizeMB.
#
# --option args outside the list of types in vals.h are treated
# as literal (non-variable) strings or numbers.
#
# positional args can be multiple types separated by |, e.g. VG|LV|Tag
#
# If a positional arg is repeatable, it is followed by ..., e.g. VG|LV|Tag ...
#
# LV can have a suffix indicating the LV type, e.g. LV_linear, LV_thinpool.
# LV_raid represents any raidN. LV_type1_type2_type3 when the LV is
# limited to multiple specific types.
#
# Note that two commands whose required paramters differ only by
# the LV types they accept are ambiguous. That is, they cannot be
# distinguished by just looking at the command, but require reading
# the VG to find the LV type. So, command definitions that differ
# only in accepted LV types are not allowed. It would be best to
# distinguish them by using different option names.
# There are FIXME's below for some of these cases.
#
# VG, LV can have the suffix _new, indicating the named VG or LV
# does not yet exist.
#
# If Select is included in pos_arg, it means that the pos_arg
# may be empty if the --select option is used.
#
# --size and --extents are interchangable, but only --size is used
# in these definitions to keep them simpler. --extents is
# automatically included and recognized as an alternative to --size.
#
# lvcreate generally requires a VG arg in position 1 and does not
# require the --name option (when --name is omitted, a name is
# generated). But, all commands of that form have a variant which
# is not defined here, but which is automatically recognized as
# being equivalent. That variant allows the required VG arg to
# be omitted when --name VG/LV is specified, or when the
# LVM_VG_NAME env var is set and --name LV is specified.
# The lvcreate variants with --name and without a VG arg are
# automatically recognized as an alternative to the defined
# command forms requiring the VG and no --name.
# Also, --thinpool VG/LV or --cachepool VG/LV can be used in
# place of --name to provide the VG name instead of pos 1.
#
# Some options have multiple names, but only one form of the name
# is used in these definitions. Synonyms will be recognized when
# matching a command to a command definition.
#
# used in definitions below (equivalent but not used in definitions)
# mirrorlog core (not corelog)
# resizeable (resizable or allocation)
# allocatable (allocation)
# activate (available)
# rebuild (raidrebuild)
# syncaction (raidsyncaction)
# writemostly (raidwritemostly)
# minrecoveryrate (raidminrecoveryrate)
# maxrecoveryrate (raidmaxrecoveryrate)
# writebehind (raidwritebehind)
# virtualsize (virtualoriginsize)
#
# metadatacopies/pvmetadatacopies/vgmetadatacopies are not considered
# synonyms for the purpose of command definitions, but each variant is
# specified when it is accepted.
#
# "---" is like a comment line, used to separate text for readability
#
# ID: A unique string identifying the command. Two commands that do
# the same thing, but are alternate syntaxes can share the same ID,
# in which case the implementation would have to sort out which
# args to look at for the required parameters. Or, the two commands
# could use differnet IDs, in which case the implementation would
# know where to look for each parameter.
#
# DESC: A description of the command.
#
#
# For efficiency, sets of options can be defined and reused
# in multiple command definitions.
#
# To define a common set of options:
# OO_NAME: --foo, --bar String
#
# To use this set of options, include it on the OO: line, e.g.
# OO: --example, OO_NAME
#
# which is expaneded to
# OO: --example, --foo, --bar String
#
# Including OO_NAME after a command name on the required line
# means that any one of the options is required and the rest
# are optional. The usage syntax for this case is printed as:
# command (--foo A, --bar B)
#
#
# OO_ALL is included in every command automatically.
# FIXME: add --force and --test to OO_ALL so that all commands will
# accept them even if they are not used?
#
OO_ALL: --commandprofile String, --config String, --debug,
--driverloaded Bool, --help, --profile String, --quiet,
--verbose, --version, --yes
#
# This list only applies to printing the usage text.
# These common options are displayed once at the end of
# a given command's usage. This is done to avoid excessive
# repetition of common options, which may obscure the more
# interesting and relevant parts of a common prototype.
# This definition is *only* used when generating the command
# usage strings, and is the basis for the division between
# the "usage" and "usage_common" strings. This OO defn does
# not relate to which optional opts are accepted by commands,
# which is defined by the OO line.
#
OO_USAGE_COMMON: OO_ALL, --force, --test, --noudevsync
#
# options for pvs, lvs, vgs, fullreport
#
OO_REPORT: --aligned, --all, --binary, --configreport String, --foreign,
--ignorelockingfailure, --ignoreskippedcluster, --logonly,
--nameprefixes, --noheadings, --nolocking, --nosuffix,
--options String, --partial, --readonly, --reportformat String, --rows,
--select String, --separator String, --shared, --sort String,
--trustcache, --unbuffered, --units Units, --unquoted
#
# options for config, dumpconfig, lvmconfig
#
OO_CONFIG: --atversion String, --typeconfig String, --file String, --ignoreadvanced,
--ignoreunsupported, --ignorelocal, --list, --mergedconfig, --metadataprofile String,
--sinceversion String, --showdeprecated, --showunsupported, --validate, --withsummary,
--withcomments, --withspaces, --unconfigured, --withversions
---
# None of these can function as a required option for lvchange.
OO_LVCHANGE: --autobackup Bool, --force, --ignorelockingfailure,
--ignoremonitoring, --ignoreskippedcluster, --noudevsync,
--reportformat String, --sysinit, --test, --select String
# Any of these can function as a required option for lvchange.
# profile is also part of OO_ALL, but is repeated in OO_LVCHANGE_META
# because it can function as a required opt.
OO_LVCHANGE_META: --addtag Tag, --deltag Tag,
--alloc Alloc, --contiguous Bool,
--detachprofile, --metadataprofile String, --profile String,
--permission Permission, --readahead Readahead, --setactivationskip Bool,
--errorwhenfull Bool, --discards Discards, --zero Bool,
--cachemode CacheMode, --cachepolicy String, --cachesettings String,
--minrecoveryrate SizeKB, --maxrecoveryrate SizeKB,
--writebehind Number, --writemostly PV
lvchange OO_LVCHANGE_META VG|LV|Tag|Select ...
OO: OO_LVCHANGE
ID: lvchange_properties
DESC: Change a general LV property.
lvchange --resync VG|LV_raid_mirror|Tag|Select ...
OO: OO_LVCHANGE
ID: lvchange_resync
DESC: Resyncronize a mirror or raid LV.
lvchange --syncaction String VG|LV_raid|Tag|Select ...
OO: OO_LVCHANGE
ID: lvchange_syncaction
DESC: Resynchronize or check a raid LV.
lvchange --rebuild PV VG|LV_raid|Tag|Select ...
OO: OO_LVCHANGE
ID: lvchange_rebuild
DESC: Reconstruct data on specific PVs of a raid LV.
lvchange --activate Active VG|LV|Tag|Select ...
OO: --activationmode ActivationMode, --partial, --ignoreactivationskip, OO_LVCHANGE_META, OO_LVCHANGE
ID: lvchange_activate
DESC: Activate or deactivate an LV.
lvchange --refresh VG|LV|Tag|Select ...
OO: --partial, OO_LVCHANGE
ID: lvchange_refresh
DESC: Reactivate an LV using the latest metadata.
lvchange --monitor Bool VG|LV|Tag|Select ...
OO: --poll Bool, OO_LVCHANGE
ID: lvchange_monitor
DESC: Start or stop monitoring an LV from dmeventd.
lvchange --poll Bool VG|LV|Tag|Select ...
OO: --monitor Bool, OO_LVCHANGE
ID: lvchange_poll
DESC: Start or stop processing an LV conversion.
lvchange --persistent Bool VG|LV|Tag|Select ...
OO: --minor Number, --major Number, OO_LVCHANGE
ID: lvchange_persistent
DESC: Make the minor device number persistent for an LV.
---
OO_LVCONVERT_RAID: --mirrors SNumber, --stripes_long Number,
--stripesize SizeKB, --regionsize SizeMB, --interval Number
OO_LVCONVERT_POOL: --poolmetadata LV, --poolmetadatasize SizeMB,
--poolmetadataspare Bool, --readahead Readahead, --chunksize SizeKB
OO_LVCONVERT: --alloc Alloc, --background, --force, --noudevsync,
--test, --usepolicies
---
# FIXME: lvconvert --merge is an extremely ambiguous command.
# It can do very different operations, but which one depends
# on knowing the LV type. So, the command doesn't know what
# it's actually doing until quite late, when processing a
# single LV.
#
# Use different option names for different merge operations
# so that we can have different command definitions,
# different behaviors, different optional options, etc:
#
# lvconvert --merge-mirror LV_linear_striped_raid ...
# DESC: Merge LV that was previously split from a mirror.
#
# lvconvert --merge-thin LV_thin
# DESC: Merge thin LV into its origin LV.
#
# lvconvert --merge-snapshot LV_snapshot
# DESC: Merge COW snapshot LV into its origin.
#
# Then we could add VG|Tag to --merge-mirror arg pos 1, because
# "lvconvert --merge VG|Tag" is a terrible command. It will do
# different operations on each LV it finds, depending on the
# current LV type.
lvconvert --merge LV_linear_striped_raid_thin_snapshot|VG|Tag ...
OO: --background, --interval Number, OO_LVCONVERT
ID: lvconvert_merge
DESC: Merge LV that was previously split from a mirror.
DESC: Merge thin LV into its origin LV.
DESC: Merge COW snapshot LV into its origin.
---
# FIXME: by using two different positional args, this is the
# single violation of the standard method of using process_each_lv().
# Before calling process_each, it steals the first positional arg
# and adjusts argv/argc so it's not seen by process_each.
lvconvert --type snapshot LV_linear_striped_raid LV_snapshot
OO: --chunksize SizeKB, --zero Bool, OO_LVCONVERT
ID: lvconvert_combine_split_snapshot
DESC: Combine LV with a previously split snapshot LV.
# alternate form of lvconvert --type snapshot
lvconvert --snapshot LV_linear_striped_raid LV_snapshot
OO: --chunksize SizeKB, --zero Bool, OO_LVCONVERT
ID: lvconvert_combine_split_snapshot
DESC: Combine LV with a previously split snapshot LV
DESC: (variant, infers --type snapshot).
---
lvconvert --type thin --thinpool LV LV_linear_striped_raid
OO: --thin, --originname LV_new, --zero Bool, OO_LVCONVERT_POOL, OO_LVCONVERT
ID: lvconvert_to_thin_with_external
DESC: Convert LV to type thin with an external origin.
# alternate form of lvconvert --type thin
lvconvert --thin --thinpool LV LV_linear_striped_raid
OO: --type thin, --originname LV_new, --zero Bool, OO_LVCONVERT_POOL, OO_LVCONVERT
ID: lvconvert_to_thin_with_external
DESC: Convert LV to type thin with an external origin
DESC: (variant, infers --type thin).
---
# FIXME: I don't think --zero applies when creating cache LV,
# but it's used in a test. Should the test be fixed and
# --zero removed here?
lvconvert --type cache --cachepool LV LV_linear_striped_raid_thinpool
OO: --cache, --cachemode CacheMode, --cachepolicy String,
--cachesettings String, --zero Bool, OO_LVCONVERT_POOL, OO_LVCONVERT
ID: lvconvert_to_cache_vol
DESC: Convert LV to type cache.
# alternate form of lvconvert --type cache
lvconvert --cache --cachepool LV LV_linear_striped_raid_thinpool
OO: --type cache, --cachemode CacheMode, --cachepolicy String,
--cachesettings String, --zero Bool, OO_LVCONVERT_POOL, OO_LVCONVERT
ID: lvconvert_to_cache_vol
DESC: Convert LV to type cache (variant, infers --type cache).
---
# FIXME: stripes is an odd option, but if the pool metadata LV
# is created by the command, that LV is created with the
# specified number of stripes. Maybe we could separate the
# command definitions for cases where pool metadata LV is
# created and limit the use of stripes to that case?
lvconvert --type thin-pool LV_linear_striped_raid_cache
OO: --stripes_long Number, --stripesize SizeKB,
--discards Discards, --zero Bool, OO_LVCONVERT_POOL, OO_LVCONVERT
ID: lvconvert_to_thinpool
DESC: Convert LV to type thin-pool.
# alternate form of lvconvert --type thin-pool
# deprecated because of non-standard syntax (missing positional arg)
lvconvert --thinpool LV_linear_striped_raid_cache
OO: --stripes_long Number, --stripesize SizeKB,
--discards Discards, --zero Bool, OO_LVCONVERT_POOL, OO_LVCONVERT
ID: lvconvert_to_thinpool
DESC: Convert LV to type thin-pool (variant, use --type thin-pool).
---
# FIXME: I don't think that --cachemode, --cachepolicy, --cachesettings
# are meant to be used when creating a cache pool, but they are used
# in one test. Should that test be fixed and these options removed?
lvconvert --type cache-pool LV_linear_striped_raid
OO: OO_LVCONVERT_POOL, OO_LVCONVERT,
--cachemode CacheMode, --cachepolicy String, --cachesettings String
ID: lvconvert_to_cachepool
DESC: Convert LV to type cache-pool.
# alternate form of lvconvert --type cache-pool
# deprecated because of non-standard syntax (missing positional arg)
lvconvert --cachepool LV_linear_striped_raid
OO: --type cache-pool, OO_LVCONVERT_POOL, OO_LVCONVERT,
--cachemode CacheMode, --cachepolicy String, --cachesettings String
ID: lvconvert_to_cachepool
DESC: Convert LV to type cache-pool (variant, use --type cache-pool).
---
# FIXME: we should be able to remove LV_mirror from the list of accepted
# LV types, but there are some dubious commands in the test suite that
# fail without it (the tests should be cleaned up to avoid using commands
# that don't make sense.)
#
# FIXME: it would be nice to remove LV_raid1 from the list of accepted
# LV types and let raid1 be covered by just the second definition, but
# unfortunatley lvconvert --type mirror --mirrors N LV_raid1 will
# match the first definition since the LV type cannot be used when
# choosing a matching command definition.
lvconvert --type mirror --mirrors SNumber LV_linear_striped_raid1_mirror
OO: --mirrorlog MirrorLog, OO_LVCONVERT_RAID, OO_LVCONVERT
OP: PV ...
ID: lvconvert_to_mirror
DESC: Convert LV to type mirror.
lvconvert --type mirror LV_raid1
OO: --mirrors SNumber, OO_LVCONVERT_RAID, OO_LVCONVERT
OP: PV ...
ID: lvconvert_to_mirror
DESC: Convert LV from type raid1 to type mirror.
---
# FIXME: by using specific raid levels, e.g. raid1, we could
# specify other required options, e.g. --mirrors. This may
# help the second fixme...
#
# FIXME: there are two different operations here, and it would
# be nice to split them into two unambiguous command lines:
#
# 1. lvconvert --type raid LV_linear_striped_mirror
# DESC: Convert LV to type raid.
#
# 2. lvconvert --type raid LV_raid
# DESC: Change LV raid type.
lvconvert --type raid LV_linear_striped_mirror_raid
OO: OO_LVCONVERT_RAID, OO_LVCONVERT
OP: PV ...
ID: lvconvert_general_to_raid
DESC: Convert LV to type raid.
DESC: Change LV between raid types.
---
# FIXME: there are two unique operations here that differ
# by only the LV type, so they have to be defined together.
# We don't know what this command is going to do until after
# the LV is read. It would be better if the alternate form
# variant were dropped to remove the ambiguity
# (then --type raid1|mirror would be required to change a
# linear or striped LV to raid1|mirror.)
#
# First command definition
# this changes the number of mirror images in a raid1|mirror LV.
# lvconvert --mirrors SNumber LV_raid_mirror
# ID: lvconvert_change_mirror_images
# DESC: Change the number of mirror images in the LV.
#
# Second command definition
# alternate form of: lvconvert --type raid1|mirror LV_linear_striped
# lvconvert --mirrors SNumber LV_linear_striped
# ID: lvconvert_to_raid1_or_mirror
# DESC: Convert LV to type raid1 or mirror
# DESC: (variant, infers --type raid1|mirror).
# first def is unique, second def is alternate form of lvconvert --type raid1|mirror
lvconvert --mirrors SNumber LV_raid_mirror_linear_striped
OO: --type raid1, --type mirror, --mirrorlog MirrorLog, OO_LVCONVERT_RAID, OO_LVCONVERT
OP: PV ...
ID: lvconvert_to_mirrored_or_change_image_count
DESC: Change the number of mirror images in a raid1 or mirror LV.
DESC: Convert a linear or striped LV to type raid1 or mirror
DESC: (variant, infers --type raid1|mirror).
---
lvconvert --type striped LV_raid
OO: OO_LVCONVERT_RAID, OO_LVCONVERT
OP: PV ...
ID: lvconvert_raid_to_striped
DESC: Convert LV to type striped.
---
lvconvert --type linear LV_raid_mirror
OO: OO_LVCONVERT
ID: lvconvert_raid_or_mirror_to_linear
DESC: Convert LV to type linear.
# FIXME: the 'mirrors 0' trick as an alias for linear
# is used inconsistently, confusing things and making
# definitions difficult.
# alternate form of lvconvert --type linear
lvconvert --mirrors 0 LV_raid_mirror
OO: --type linear, --type mirror, OO_LVCONVERT
ID: lvconvert_raid_or_mirror_to_linear
DESC: Convert LV to type linear (variant, infers --type linear).
---
lvconvert --splitmirrors Number --name LV_new LV_raid1_mirror_cache
OO: OO_LVCONVERT
OP: PV ...
ID: lvconvert_split_mirror_images_to_new
DESC: Split images from a raid1 or mirror LV and use them to create a new LV.
lvconvert --splitmirrors Number --trackchanges LV_raid1_cache
OO: OO_LVCONVERT
OP: PV ...
ID: lvconvert_split_mirror_images_and_track
DESC: Split images from a raid1 LV and track changes to origin.
---
# FIXME: use specific option names to distinguish these two
# very different commands, e.g.
#
# lvconvert --repair-pvs LV_raid_mirror
# DESC: Replace failed PVs in a raid or mirror LV.
#
# lvconvert --repair-thinpool LV_thinpool
# DESC: Repair a thin pool.
#
# lvm may want to do different things, or allow different options
# depending on which operation is being run, but as it stands, it
# cannot do anything operation-specific until after the VG is read
# and the LV type is known.
lvconvert --repair LV_raid_mirror_thinpool
OO: OO_LVCONVERT
ID: lvconvert_repair_pvs_or_thinpool
DESC: Replace failed PVs in a raid or mirror LV.
DESC: Repair a thin pool.
---
lvconvert --replace PV LV_raid
OO: OO_LVCONVERT
OP: PV ...
ID: lvconvert_replace_pv
DESC: Replace specific PV(s) in a raid* LV with another PV.
---
lvconvert --mirrorlog MirrorLog LV_mirror
OO: OO_LVCONVERT
OP: PV ...
ID: lvconvert_change_mirrorlog
DESC: Change the type of mirror log used by a mirror LV.
---
lvconvert --splitcache LV_cachepool_cache_thinpool
OO: OO_LVCONVERT
ID: lvconvert_split_and_keep_cachepool
DESC: Separate and keep the cache pool from a cache LV.
---
lvconvert --uncache LV_cache_thinpool
OO: OO_LVCONVERT
ID: lvconvert_split_and_delete_cachepool
DESC: Separate and delete the cache pool from a cache LV.
---
lvconvert --splitsnapshot LV_snapshot
OO: OO_LVCONVERT
ID: lvconvert_split_cow_snapshot
DESC: Separate a COW snapshot from its origin LV.
---
# FIXME: add a new option defining this operation, e.g. --poll-mirror
# The purpose of this command is not entirely clear.
lvconvert LV_mirror
OO: OO_LVCONVERT
ID: lvconvert_poll_mirror
DESC: Poll mirror LV to collapse resync layers.
---
# FIXME: add a new option defining this operation, e.g. --swapmetadata
lvconvert --poolmetadata LV LV_thinpool_cachepool
OO: OO_LVCONVERT
ID: lvconvert_swap_pool_metadata
DESC: Swap metadata LV in a thin pool or cache pool (temporary command).
---
# --extents is not specified; it's an automatic alternative for --size
OO_LVCREATE: --addtag Tag, --alloc Alloc, --autobackup Bool, --activate Active,
--contiguous Bool, --ignoreactivationskip, --ignoremonitoring, --major Number,
--metadataprofile String, --minor Number, --monitor Bool, --name String, --nosync,
--noudevsync, --permission Permission, --persistent Bool, --readahead Readahead,
--reportformat String, --setactivationskip Bool, --test, --wipesignatures Bool,
--zero Bool
OO_LVCREATE_CACHE: --cachemode CacheMode, --cachepolicy String, --cachesettings String
OO_LVCREATE_POOL: --poolmetadatasize SizeMB, --poolmetadataspare Bool, --chunksize SizeKB
# FIXME: it's silly to include --mirrors 0 here. Fix the tests to not use
# --mirrors 0 in commands that do not accept any non-zero --mirrors
# option, and then remove this. Accepting an option, only so that the
# option's value can invalidate the use of the option is not advisable.
OO_LVCREATE_THIN: --discards Discards, --errorwhenfull Bool, --mirrors 0
OO_LVCREATE_RAID: --mirrors SNumber, --stripes Number, --stripesize SizeKB,
--regionsize SizeMB, --minrecoveryrate SizeKB, --maxrecoveryrate SizeKB
---
lvcreate --type error --size SizeMB VG
OO: OO_LVCREATE
ID: lvcreate_error_vol
DESC: Create an LV that returns errors when used.
---
lvcreate --type zero --size SizeMB VG
OO: OO_LVCREATE
ID: lvcreate_zero_vol
DESC: Create an LV that returns zeros when read.
---
# FIXME: consider removing the --mirrors 0, --stripes 1 options
# and just reporting an error (or ignoring) if mirrors or stripes
# options are given. Same reasoning as above: it's confusing to
# advertise an option when the only value accepted for the option
# makes the option do nothing.
lvcreate --type linear --size SizeMB VG
OO: --mirrors 0, --stripes 1, OO_LVCREATE
OP: PV ...
ID: lvcreate_linear
DESC: Create a linear LV.
# This is the one place we mention the optional --name
# because it's the most common case and may be confusing
# to people to not see the name parameter.
lvcreate --size SizeMB VG
OO: --type linear, --mirrors 0, --stripes 1, OO_LVCREATE
OP: PV ...
ID: lvcreate_linear
DESC: Create a linear LV (default --type linear).
DESC: When --name is omitted, the name is generated.
---
lvcreate --type striped --size SizeMB VG
OO: --stripes Number, --stripesize SizeKB, OO_LVCREATE
OP: PV ...
ID: lvcreate_striped
DESC: Create a striped LV.
lvcreate --stripes Number --size SizeMB VG
OO: --type striped, --stripesize SizeKB, OO_LVCREATE
OP: PV ...
ID: lvcreate_striped
DESC: Create a striped LV (infers --type striped).
---
lvcreate --type mirror --size SizeMB VG
OO: --mirrors SNumber, --mirrorlog MirrorLog, --regionsize SizeMB, --stripes Number, OO_LVCREATE
OP: PV ...
ID: lvcreate_mirror
DESC: Create a mirror LV (also see --type raid1).
# alternate form of lvcreate --type raid1|mirror
lvcreate --mirrors SNumber --size SizeMB VG
OO: --type raid1, --type mirror, --mirrorlog MirrorLog, --stripes Number, OO_LVCREATE_RAID, OO_LVCREATE
OP: PV ...
ID: lvcreate_mirror_or_raid1
DESC: Create a raid1 or mirror LV (variant, infers --type raid1|mirror).
---
# FIXME: we should probably expand this from raid* to
# various specific raid levels, then narrow the range
# of options applicable to each.
lvcreate --type raid --size SizeMB VG
OO: OO_LVCREATE_RAID, OO_LVCREATE
OP: PV ...
ID: lvcreate_raid_any
DESC: Create a raid LV (a specific raid level must be used, e.g. raid1.)
---
# FIXME: the LV created by these commands actually has type linear or striped,
# not snapshot as specified by the command. If LVs never have type
# snapshot, perhaps "snapshot" should not be considered an LV type, but
# another new LV property?
#
# Or, perhaps we should not promote --type snapshot, and
# prefer --snapshot as the primary syntax, since the type
# is not really snapshot (which makes it different from
# all the other cases of using --type, where the resulting
# LV does have the specified type.)
lvcreate --type snapshot --size SizeMB LV
OO: --snapshot, --stripes Number, --stripesize SizeKB,
--chunksize SizeKB, OO_LVCREATE
OP: PV ...
ID: lvcreate_cow_snapshot
DESC: Create a COW snapshot LV from an origin LV.
# alternate form of lvcreate --type snapshot
lvcreate --snapshot --size SizeMB LV
OO: --type snapshot, --stripes Number, --stripesize SizeKB,
--chunksize SizeKB, OO_LVCREATE
OP: PV ...
ID: lvcreate_cow_snapshot
DESC: Create a COW snapshot LV from an origin LV
DESC: (infers --type snapshot).
---
lvcreate --type snapshot --size SizeMB --virtualsize SizeMB VG
OO: --snapshot, --chunksize SizeKB, OO_LVCREATE
OP: PV ...
ID: lvcreate_cow_snapshot_with_virtual_origin
DESC: Create a sparse COW snapshot LV of a virtual origin LV.
# alternate form of lvcreate --type snapshot
lvcreate --snapshot --size SizeMB --virtualsize SizeMB VG
OO: --type snapshot, --chunksize SizeKB, OO_LVCREATE
OP: PV ...
ID: lvcreate_cow_snapshot_with_virtual_origin
DESC: Create a sparse COW snapshot LV of a virtual origin LV
DESC: (infers --type snapshot).
---
lvcreate --type thin-pool --size SizeMB VG
OO: --thinpool LV_new, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thinpool
DESC: Create a thin pool.
# alternate form of lvcreate --type thin-pool
lvcreate --thin --size SizeMB VG
OO: --type thin-pool, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thinpool
DESC: Create a thin pool (variant, infers --type thin-pool).
# alternate form of lvcreate --type thin-pool
lvcreate --size SizeMB --thinpool LV_new VG
OO: --thin, --type thin-pool, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thinpool
DESC: Create a thin pool named by the --thinpool arg
DESC: (variant, infers --type thin-pool).
---
lvcreate --type cache-pool --size SizeMB VG
OO: OO_LVCREATE_POOL, OO_LVCREATE_CACHE, OO_LVCREATE
OP: PV ...
ID: lvcreate_cachepool
DESC: Create a cache pool.
# alternate form of lvcreate --type cache-pool
lvcreate --cache --size SizeMB VG
OO: --type cache-pool, OO_LVCREATE_POOL, OO_LVCREATE_CACHE, OO_LVCREATE
OP: PV ...
ID: lvcreate_cachepool
DESC: Create a cache pool (variant, infers --type cache-pool).
# alternate form of lvcreate --type cache-pool
lvcreate --type cache-pool --size SizeMB --cachepool LV_new VG
OO: OO_LVCREATE_POOL, OO_LVCREATE_CACHE, OO_LVCREATE
OP: PV ...
ID: lvcreate_cachepool
DESC: Create a cache pool named by the --cachepool arg
DESC: (variant, uses --cachepool in place of --name).
---
lvcreate --type thin --virtualsize SizeMB --thinpool LV_thinpool VG
OO: --thin, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_vol
DESC: Create a thin LV in a thin pool.
# alternate form of lvcreate --type thin
lvcreate --type thin --virtualsize SizeMB LV_thinpool
OO: --thin, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_vol
DESC: Create a thin LV in a thin pool named in the first arg
DESC: (variant, also see --thinpool for naming pool).
# alternate form of lvcreate --type thin
lvcreate --virtualsize SizeMB --thinpool LV_thinpool
OO: --type thin, --thin, OO_LVCREATE_THIN, OO_LVCREATE
OP: VG
ID: lvcreate_thin_vol
DESC: Create a thin LV in a thin pool (variant, infers --type thin).
# alternate form of lvcreate --type thin
lvcreate --virtualsize SizeMB LV_thinpool
OO: --type thin, --thin, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_vol
DESC: Create a thin LV in the thin pool named in the first arg
DESC: (variant, infers --type thin, also see --thinpool for
DESC: naming pool.)
---
lvcreate --type thin LV_thin
OO: --thin, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_snapshot
DESC: Create a thin LV that is a snapshot of an existing thin LV.
# alternate form of lvcreate --type thin
lvcreate --thin LV_thin
OO: --type thin, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_snapshot
DESC: Create a thin LV that is a snapshot of an existing thin LV
DESC: (infers --type thin).
# alternate form of lvcreate --type thin
lvcreate --snapshot LV_thin
OO: --type thin, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_snapshot
DESC: Create a thin LV that is a snapshot of an existing thin LV
DESC: (infers --type thin).
lvcreate --type thin --thinpool LV_thinpool LV
OO: OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_snapshot_of_external
DESC: Create a thin LV that is a snapshot of an external origin LV.
# alternate form of lvcreate --type thin --thinpool LV_thinpool LV
lvcreate --snapshot --thinpool LV_thinpool LV
OO: --type thin, OO_LVCREATE_THIN, OO_LVCREATE
ID: lvcreate_thin_snapshot_of_external
DESC: Create a thin LV that is a snapshot of an external origin LV
DESC: (infers --type thin).
---
# stripes option is not intuitive when creating a thin LV,
# but here it applies to creating the new thin pool that
# is used for the thin LV
lvcreate --type thin --virtualsize SizeMB --size SizeMB --thinpool LV_new
OO: OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thin_vol_and_thinpool
DESC: Create a thin LV, first creating a thin pool for it,
DESC: where the new thin pool is named by the --thinpool arg.
# alternate form of lvcreate --type thin
lvcreate --thin --virtualsize SizeMB --size SizeMB --thinpool LV_new
OO: --type thin, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thin_vol_and_thinpool
DESC: Create a thin LV, first creating a thin pool for it,
DESC: where the new thin pool is named by the --thinpool arg
DESC: (variant, infers --type thin).
lvcreate --type thin --virtualsize SizeMB --size SizeMB LV_new
OO: OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thin_vol_and_thinpool
DESC: Create a thin LV, first creating a thin pool for it,
DESC: where the new thin pool is named in the first arg.
lvcreate --thin --virtualsize SizeMB --size SizeMB LV_new
OO: --type thin, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thin_vol_and_thinpool
DESC: Create a thin LV, first creating a thin pool for it,
DESC: where the new thin pool is named in the first arg
DESC: (variant, infers --type thin).
lvcreate --type thin --virtualsize SizeMB --size SizeMB VG
OO: OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thin_vol_and_thinpool
DESC: Create a thin LV, first creating a thin pool for it,
DESC: where the new thin pool name is generated automatically.
---
lvcreate --size SizeMB --virtualsize SizeMB VG
OO: --type thin, --type snapshot, --thin, --snapshot, OO_LVCREATE_POOL, OO_LVCREATE_THIN, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_thin_vol_with_thinpool_or_sparse_snapshot
DESC: Create a thin LV, first creating a thin pool for it
DESC: (infers --type thin).
DESC: Create a sparse snapshot of a virtual origin LV
DESC: (infers --type snapshot).
DESC: Chooses --type thin or --type snapshot according to
DESC: config setting sparse_segtype_default.
---
# stripes option is not intuitive when creating a cache LV,
# but here it applies to creating the new origin that
# is used to create the cache LV
lvcreate --type cache --size SizeMB --cachepool LV_cachepool VG
OO: --cache, OO_LVCREATE_POOL, OO_LVCREATE_CACHE, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_cache_vol_with_new_origin
DESC: Create a cache LV, first creating a new origin LV,
DESC: then combining it with the existing cache pool named
DESC: by the --cachepool arg.
# alternate form of lvcreate --type cache
lvcreate --size SizeMB --cachepool LV_cachepool VG
OO: --type cache, --cache, OO_LVCREATE_CACHE, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_cache_vol_with_new_origin
DESC: Create a cache LV, first creating a new origin LV,
DESC: then combining it with the existing cache pool named
DESC: by the --cachepool arg (variant, infers --type cache).
# alternate form of lvcreate --type cache
lvcreate --type cache --size SizeMB LV_cachepool
OO: --cache, OO_LVCREATE_POOL, OO_LVCREATE_CACHE, OO_LVCREATE,
--stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_cache_vol_with_new_origin
DESC: Create a cache LV, first creating a new origin LV,
DESC: then combining it with the existing cache pool named
DESC: in the first arg (variant, also use --cachepool).
# This is a ridiculously crazy command which nobody could
# understand. It should be be eliminated. It does two different
# things depending on whether LV in pos 1 is a cachepool LV
# or not. Both variations are unnecessary.
#
# 1. If LV is a cachepool, then it's an alternate form of
# an already complicated command above.
#
# # alternate form for lvcreate_cache_vol_with_new_origin
# lvcreate --cache --size SizeMB LV_cachepool
# OO: --type cache, --cache, OO_LVCREATE_CACHE, OO_LVCREATE, --stripes Number, --stripesize SizeKB
# OP: PV ...
# ID: lvcreate_cache_vol_with_new_origin
# DESC: Create a cache LV, first creating a new origin LV,
# DESC: then combining it with the existing cache pool named
# DESC: in the first arg (variant, infers --type cache,
# DESC: also use --cachepool).
#
# 2. If LV is not a cachepool, then it's a disguised lvconvert.
#
# # FIXME: this should be done by lvconvert, and this command removed
# lvcreate --type cache --size SizeMB LV
# OO: OO_LVCREATE_POOL, OO_LVCREATE_CACHE, OO_LVCREATE
# OP: PV ...
# ID: lvcreate_convert_to_cache_vol_with_cachepool
# DESC: Convert the specified LV to type cache after creating a new
# DESC: cache pool LV to use (use lvconvert).
#
# Note that stripes are accepted by the first and not by the
# second, but it's not possible to validate this until after
# the LV type is known.
#
# So, to define this syntax we have to combine both of
# those variants, each crazy on it's own, into one
# ridiculous command.
# def1: alternate form of lvcreate --type cache, or
# def2: it should be done by lvconvert.
lvcreate --cache --size SizeMB LV
OO: OO_LVCREATE_CACHE, OO_LVCREATE, --stripes Number, --stripesize SizeKB
OP: PV ...
ID: lvcreate_cache_vol_with_new_origin_or_convert_to_cache_vol_with_cachepool
DESC: When LV is a cache pool, create a cache LV,
DESC: first creating a new origin LV, then combining it with
DESC: the existing cache pool named in the first arg
DESC: (variant, infers --type cache, also use --cachepool).
DESC: When LV is not a cache pool, convert the specified LV
DESC: to type cache after creating a new cache pool LV to use
DESC: (use lvconvert).
---
lvdisplay
OO: --aligned, --all, --binary, --colon, --columns,
--configreport String, --foreign, --history, --ignorelockingfailure,
--ignoreskippedcluster, --logonly, --maps, --noheadings,
--nosuffix, --options String, --sort String, --partial, --readonly,
--reportformat String, --segments, --select String, --separator String,
--shared, --unbuffered, --units Units
OP: VG|LV|Tag ...
ID: lvdisplay_general
---
# --extents is not specified; it's an automatic alternative for --size
lvextend --size SizeMB LV
OO: --alloc Alloc, --autobackup Bool, --force, --mirrors SNumber,
--nofsck, --nosync, --noudevsync, --reportformat String, --resizefs,
--stripes Number, --stripesize SizeKB, --test, --poolmetadatasize SizeMB
OP: PV ...
ID: lvextend_by_size
DESC: Extend an LV by a specified size.
lvextend LV PV ...
OO: --alloc Alloc, --autobackup Bool, --force, --mirrors SNumber,
--nofsck, --nosync, --noudevsync,
--reportformat String, --resizefs, --stripes Number, --stripesize SizeKB,
--test
ID: lvextend_by_pv
DESC: Extend an LV by a specified PV.
lvextend --poolmetadatasize SizeMB LV_thinpool
OO: --alloc Alloc, --autobackup Bool, --force, --mirrors SNumber,
--nofsck, --nosync, --noudevsync,
--reportformat String, --stripes Number, --stripesize SizeKB,
--test
OP: PV ...
ID: lvextend_pool_metadata_by_size
DESC: Extend a pool metadata SubLV by a specified size.
lvextend --usepolicies LV_thinpool_snapshot
OO: --alloc Alloc, --autobackup Bool, --force, --mirrors SNumber,
--nofsck, --nosync, --noudevsync,
--reportformat String, --resizefs,
--test
ID: lvextend_by_policy
DESC: Extend an LV according to a predefined policy.
---
lvmconfig
OO: OO_CONFIG
ID: lvmconfig_general
---
lvreduce --size SizeMB LV
OO: --autobackup Bool, --force, --nofsck, --noudevsync,
--reportformat String, --resizefs, --test
ID: lvreduce_general
---
lvremove VG|LV|Tag|Select ...
OO: --autobackup Bool, --force, --nohistory, --noudevsync,
--reportformat String, --select String, --test
ID: lvremove_general
---
lvrename VG LV LV_new
OO: --autobackup Bool, --noudevsync, --reportformat String, --test
ID: lvrename_vg_lv_lv
lvrename LV LV_new
OO: --autobackup Bool, --noudevsync, --reportformat String, --test
ID: lvrename_lv_lv
---
lvresize --size SizeMB LV
OO: --alloc Alloc, --autobackup Bool, --force,
--nofsck, --nosync, --noudevsync, --reportformat String, --resizefs,
--stripes Number, --stripesize SizeKB, --test, --poolmetadatasize SizeMB
OP: PV ...
ID: lvresize_by_size
DESC: Resize an LV by a specified size.
lvresize LV PV ...
OO: --alloc Alloc, --autobackup Bool, --force,
--nofsck, --nosync, --noudevsync,
--reportformat String, --resizefs, --stripes Number, --stripesize SizeKB,
--test
ID: lvresize_by_pv
DESC: Resize an LV by a specified PV.
lvresize --poolmetadatasize SizeMB LV_thinpool
OO: --alloc Alloc, --autobackup Bool, --force,
--nofsck, --nosync, --noudevsync,
--reportformat String, --stripes Number, --stripesize SizeKB,
--test
OP: PV ...
ID: lvresize_pool_metadata_by_size
DESC: Resize a pool metadata SubLV by a specified size.
---
lvs
OO: --history, --segments, OO_REPORT
OP: VG|LV|Tag ...
ID: lvs_general
---
lvscan
OO: --all, --blockdevice, --ignorelockingfailure, --partial,
--readonly, --reportformat String, --cache_long
ID: lvscan_general
---
# None of these can function as a required option for pvchange.
OO_PVCHANGE: --autobackup Bool, --force, --ignoreskippedcluster,
--reportformat String, --test, --uuid
# Any of these can function as a required option for pvchange.
OO_PVCHANGE_META: --allocatable Bool, --addtag Tag, --deltag Tag,
--uuid, --metadataignore Bool
pvchange OO_PVCHANGE_META --all
OO: OO_PVCHANGE
ID: pvchange_properties_all
DESC: Change properties of all PVs.
pvchange OO_PVCHANGE_META PV|Select ...
OO: --select String, OO_PVCHANGE
ID: pvchange_properties_some
DESC: Change properties of specified PVs.
---
pvresize PV ...
OO: --setphysicalvolumesize SizeMB, --reportformat String, --test
ID: pvresize_general
---
pvck PV ...
OO: --labelsector Number
ID: pvck_general
---
# Use --uuidstr here which will be converted to uuidstr_ARG
# which is actually --uuid string on the command line.
pvcreate PV ...
OO: --dataalignment SizeKB, --dataalignmentoffset SizeKB, --bootloaderareasize SizeMB,
--force, --test, --labelsector Number, --metadatatype MetadataType,
--pvmetadatacopies Number, --metadatacopies MetadataCopies, --metadatasize SizeMB,
--metadataignore Bool, --norestorefile, --setphysicalvolumesize SizeMB,
--reportformat String, --restorefile String, --uuidstr String, --zero Bool
ID: pvcreate_general
---
pvdisplay
OO: --aligned, --all, --binary, --colon, --columns, --configreport String,
--foreign, --ignorelockingfailure, --ignoreskippedcluster,
--logonly, --maps, --noheadings, --nosuffix, --options String,
--readonly, --reportformat String, --select String, --separator String, --shared,
--short, --sort String, --unbuffered, --units Units
OP: PV|Tag ...
ID: pvdisplay_general
---
pvmove PV
OO: --abort, --alloc Alloc, --atomic, --autobackup Bool, --background,
--interval Number, --name LV, --noudevsync, --reportformat String, --test
OP: PV ...
ID: pvmove_one
DESC: Move PV extents.
pvmove
OO: --abort, --background, --test
ID: pvmove_any
DESC: Continue or abort existing pvmove operations.
---
pvremove PV ...
OO: --force, --reportformat String, --test
ID: pvremove_general
---
pvs
OO: --segments, OO_REPORT
OP: PV|Tag ...
ID: pvs_general
---
pvscan
OO: --ignorelockingfailure, --reportformat String, --exported, --novolumegroup,
--short, --uuid
ID: pvscan_show
DESC: Display PV information.
pvscan --cache_long
OO: --ignorelockingfailure, --reportformat String, --background,
--activate Active, --major Number, --minor Number,
OP: PV|String ...
ID: pvscan_cache
DESC: Populate the lvmetad cache by scanning PVs.
---
vgcfgbackup
OO: --file String, --foreign, --ignorelockingfailure, --partial, --readonly,
--reportformat String
OP: VG ...
ID: vgcfgbackup_general
---
vgcfgrestore VG
OO: --file String, --force_long, --list, --metadatatype MetadataType, --test
ID: vgcfgrestore_by_vg
vgcfgrestore --list --file String
ID: vgcfgrestore_by_file
---
# None of these can function as a required option for vgchange.
OO_VGCHANGE: --autobackup Bool, --ignoremonitoring, --ignoreskippedcluster,
--noudevsync, --reportformat String, --select String, --test, --force
# Any of these can function as a required option for vgchange.
# profile is also part of OO_ALL, but is repeated in OO_VGCHANGE_META
# because it can function as a required opt.
OO_VGCHANGE_META: --addtag Tag, --deltag Tag,
--logicalvolume Number, --maxphysicalvolumes Number, --alloc Alloc, --uuid,
--clustered Bool, --metadatacopies MetadataCopies, --vgmetadatacopies MetadataCopies,
--physicalextentsize SizeMB, --resizeable Bool, --systemid String, --locktype LockType,
--profile String, --detachprofile, --metadataprofile String,
vgchange OO_VGCHANGE_META
OO: OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_properties
DESC: Change a general VG property.
vgchange --monitor Bool
OO: --sysinit, --ignorelockingfailure, --poll Bool, OO_VGCHANGE_META, OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_monitor
DESC: Start or stop monitoring LVs from dmeventd.
vgchange --poll Bool
OO: --ignorelockingfailure, OO_VGCHANGE_META, OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_poll
DESC: Start or stop processing LV conversions.
vgchange --activate Active
OO: --activationmode ActivationMode, --ignoreactivationskip, --partial, --sysinit,
--ignorelockingfailure, --monitor Bool, --poll Bool, OO_VGCHANGE_META, OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_activate
DESC: Activate or deactivate LVs.
vgchange --refresh
OO: --sysinit, --ignorelockingfailure, --monitor Bool, --poll Bool, OO_VGCHANGE_META, OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_refresh
DESC: Reactivate LVs using the latest metadata.
vgchange --lockstart
OO: --lockopt String, OO_VGCHANGE_META, OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_lockstart
DESC: Start the lockspace of a shared VG in lvmlockd.
vgchange --lockstop
OO: --lockopt String, OO_VGCHANGE_META, OO_VGCHANGE
OP: VG|Tag ...
ID: vgchange_lockstop
DESC: Stop the lockspace of a shared VG in lvmlockd.
---
vgck
OO: --reportformat String
OP: VG|Tag ...
ID: vgck_general
---
vgconvert VG ...
OO: --force, --test, --labelsector Number, --bootloaderareasize SizeMB,
--metadatatype MetadataType, --pvmetadatacopies Number,
--metadatasize SizeMB, --reportformat String
ID: vgconvert_general
---
vgcreate VG_new PV ...
OO: --addtag Tag, --alloc Alloc, --autobackup Bool, --clustered Bool, --maxlogicalvolumes Number,
--maxphysicalvolumes Number, --metadataprofile String, --metadatatype MetadataType,
--physicalextentsize SizeMB, --test, --force, --zero Bool, --labelsector Number,
--metadatasize SizeMB, --pvmetadatacopies Number, --reportformat String, --metadatacopies MetadataCopies,
--vgmetadatacopies MetadataCopies, --dataalignment SizeKB, --dataalignmentoffset SizeKB,
--shared, --systemid String, --locktype LockType, --lockopt String
ID: vgcreate_general
---
vgdisplay
OO: --activevolumegroups, --aligned, --binary, --colon, --columns,
--configreport String, --foreign, --ignorelockingfailure,
--ignoreskippedcluster, --logonly, --noheadings, --nosuffix,
--options String, --partial, --readonly, --reportformat String, --select String,
--shared, --short, --separator String, --sort String, --unbuffered, --units Units
OP: VG|Tag ...
ID: vgdisplay_general
---
OO_VGEXPORT: --reportformat String, --test
vgexport VG|Tag|Select ...
OO: --select String, OO_VGEXPORT
ID: vgexport_some
DESC: Export specified VGs.
vgexport --all
OO: OO_VGEXPORT
ID: vgexport_all
DESC: Export all VGs.
---
vgextend VG PV ...
OO: --autobackup Bool, --test,
--force, --zero Bool, --labelsector Number, --metadatatype MetadataType,
--metadatasize SizeMB, --pvmetadatacopies Number,
--metadataignore Bool, --dataalignment SizeKB, --dataalignmentoffset SizeKB,
--reportformat String, --restoremissing
ID: vgextend_general
---
OO_VGIMPORT: --force, --reportformat String, --test
vgimport VG|Tag|Select ...
OO: --select String, OO_VGIMPORT
ID: vgimport_some
DESC: Import specified VGs.
vgimport --all
OO: OO_VGIMPORT
ID: vgimport_all
DESC: Import all VGs.
---
vgimportclone PV ...
OO: --basevgname VG, --test, --import
ID: vgimportclone_general
---
vgmerge VG VG
OO: --autobackup Bool, --list, --test
ID: vgmerge_general
---
vgmknodes
OO: --ignorelockingfailure, --refresh, --reportformat String
OP: VG|LV|Tag ...
ID: vgmknodes_general
---
OO_VGREDUCE: --autobackup Bool, --force, --reportformat String, --test
vgreduce VG PV ...
OO: OO_VGREDUCE
ID: vgreduce_by_pv
DESC: Remove a PV from a VG.
vgreduce --all VG
OO: OO_VGREDUCE
ID: vgreduce_all
DESC: Remove all unused PVs from a VG.
vgreduce --removemissing VG
OO: --mirrorsonly, OO_VGREDUCE
ID: vgreduce_missing
DESC: Remove all missing PVs from a VG.
---
vgremove VG|Tag|Select ...
OO: --force, --noudevsync, --reportformat String, --select String, --test
ID: vgremove_general
---
vgrename VG VG_new
OO: --autobackup Bool, --force, --reportformat String, --test
ID: vgrename_by_name
DESC: Rename a VG.
vgrename String VG_new
OO: --autobackup Bool, --force, --reportformat String, --test
ID: vgrename_by_uuid
DESC: Rename a VG by specifying the VG UUID.
---
vgs
OO: OO_REPORT
OP: VG|Tag ...
ID: vgs_general
---
vgscan
OO: --cache_long, --ignorelockingfailure, --mknodes, --notifydbus,
--partial, --reportformat String
ID: vgscan_general
---
OO_VGSPLIT: --autobackup Bool, --test
OO_VGSPLIT_NEW: --alloc Alloc, --clustered Bool,
--maxlogicalvolumes Number, --maxphysicalvolumes Number,
--metadatatype MetadataType, --vgmetadatacopies MetadataCopies
# FIXME: it would be nice to have separate definitions
# for the cases where the destination VG exists or is new,
# but when choosing the command definition, we don't yet
# know if the destination VG exists or not. So to do this
# we'd need a command option to specify if the VG is new.
#
# this won't work:
# vgsplit VG VG PV
# ID: vgsplit_by_pv_to_existing
#
# vgsplit VG VG_new PV ...
# ID: vgsplit_by_pv_to_new
#
# but this could:
# vgsplit VG VG PV
# ID: vgsplit_by_pv_to_existing
#
# vgsplit --create-new VG VG_new PV ...
# ID: vgsplit_by_pv_to_new
#
# Then the OO_VGSPLIT_NEW options could be
# included in the second case but not the first.
vgsplit VG VG PV ...
OO: OO_VGSPLIT, OO_VGSPLIT_NEW
ID: vgsplit_by_pv
DESC: Split a VG by specified PVs.
vgsplit --name LV VG VG
OO: OO_VGSPLIT, OO_VGSPLIT_NEW
ID: vgsplit_by_lv
DESC: Split a VG by PVs in a specified LV.
---
# built-in and deprecated commands
config
OO: OO_CONFIG
OP: String ...
ID: lvmconfig_general
dumpconfig
OO: OO_CONFIG
OP: String ...
ID: lvmconfig_general
devtypes
OO: --aligned, --binary, --nameprefixes, --noheadings,
--nosuffix, --options String, --reportformat String, --rows,
--select String, --separator String, --sort String, --unbuffered, --unquoted
ID: devtypes_general
fullreport
OO: OO_REPORT
OP: VG ...
ID: fullreport_general
lastlog
OO: --reportformat String, --select String
ID: lastlog_general
lvpoll --polloperation String LV ...
OO: --abort, --autobackup Bool, --handlemissingpvs, --interval Number, --test
ID: lvpoll_general
formats
ID: formats_general
help
ID: help_general
version
ID: version_general
pvdata
ID: pvdata_general
segtypes
ID: segtypes_general
systemid
ID: systemid_general
tags
ID: tags_general
lvmchange
ID: lvmchange_general
lvmdiskscan
OO: --lvmpartition, --readonly
ID: lvmdiskscan_general
lvmsadc
ID: lvmsadc_general
lvmsar
OO: --full, --stdin
ID: lvmsar_general
View Git Blame Copy Permalink