Gitweb: https://sourceware.org/git/?p=lvm2.git;a=commitdiff;h=d6ddacf6a1543ba690fc3…
Commit: d6ddacf6a1543ba690fc349fad4c0c05cad1183b
Parent: 0ed087aae504e419ff9712ce8479b04beec38e03
Author: Heinz Mauelshagen <heinzm(a)redhat.com>
AuthorDate: Fri Mar 31 13:51:39 2017 +0200
Committer: Heinz Mauelshagen <heinzm(a)redhat.com>
CommitterDate: Fri Mar 31 13:51:39 2017 +0200
lvchange: persistent major/minor processing regression
Fix a regression introduced by new command line processing
(see starting commit 1e2420bca85da9a37570871cd70192e9ae831786)
causing activation to be rejected in combination with setting
persistent minor device numbers.
Resolves: rhbz1437611
---
tools/command-lines.in | 9 +++++++--
1 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/tools/command-lines.in b/tools/command-lines.in
index 6a73bf8..3e3f480 100644
--- a/tools/command-lines.in
+++ b/tools/command-lines.in
@@ -296,12 +296,17 @@ OO: --monitor Bool, OO_LVCHANGE
ID: lvchange_poll
DESC: Start or stop processing an LV conversion.
-lvchange --persistent y --minor Number LV
-OO: --major Number, OO_LVCHANGE
+lvchange --persistent y --minor Number --activate Active LV
+OO: --minor Number, OO_LVCHANGE
ID: lvchange_persistent
DESC: Make the minor device number persistent for an LV.
RULE: all not LV_thinpool LV_cachepool
+lvchange --persistent n --activate LV
+ID: lvchange_persistent
+DESC: Remove the minor device number persistent for an LV.
+RULE: all not LV_thinpool LV_cachepool
+
---
OO_LVCONVERT_POOL: --poolmetadata LV, --poolmetadatasize SizeMB,
Gitweb: https://sourceware.org/git/?p=lvm2.git;a=commitdiff;h=60043ae1e3ae51523c4df…
Commit: 60043ae1e3ae51523c4dfeb6ba244cfaefc98d45
Parent: 0000000000000000000000000000000000000000
Author: Alasdair G Kergon <agk(a)redhat.com>
AuthorDate: 2017-03-28 15:12 +0000
Committer: Alasdair G Kergon <agk(a)redhat.com>
CommitterDate: 2017-03-28 15:12 +0000
annotated tag: v2_02_169 has been created
at 60043ae1e3ae51523c4dfeb6ba244cfaefc98d45 (tag)
tagging 396377bc03a9f8a13f5d851306f7d2a622a1a74e (commit)
replaces v2_02_168
Release 2.02.169
This development release contains a substantial amount of change and we
recommend you use it only for testing at this stage. Please wait for
follow-on releases before running it on live systems.
The syntax of every valid command has been encoded into the same format.
Given the large number of combinations, mistakes are possible, so please
check any of the more unusual commands that you use and let us know if you
discover a command that no longer behaves the way you expected.
Most man pages are now generated in part from these new definitions and
command help information has similarly changed completely. This content
is still undergoing review and refinement.
The code also supports the new dm-cache metadata format 2 that improves
the performance and adaptation to workload changes, and the raid code
also supports several new formats and conversions.
446 files changed, 60538 insertions(+), 25701 deletions(-)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)
iEYEABECAAYFAljafXgACgkQIoGRwVZ+LBd8MQCfctT8HTr9eYSYzWbInbHJZvK0
CPsAn1La2ZKFFfvQOfLFL0zTcnIvA6ab
=igsB
-----END PGP SIGNATURE-----
Alasdair G Kergon (28):
post-release
devices: Recognise Veritas Dynamic Multipathing
Makefile: all_man also requires man-generator now
dmsetup: Indicate device repetition in usage summaries.
git: Upstream repository moved to sourceware.org
tools: Fix overriding of current_settings.
man: Revise internal man page generation process.
man: Adjust some wording and update examples style.
make: Support for 'make -R'.
man: specify man-generator path
man: reinstate lvmchange man page for now
man: Remove blank lines from pregenerated files
lvmchange: Move to a built-in command.
lvmsar/lvmsadc: Mention dmstats when run.
man: Reinstate previous lvmsar and lvmsadc pages.
man: Include lvmchange/lvmsar/lvmsadc content in lvm.8
tools: Drop unused e2fsadm refs (replaced by fsadm).
tools: Avoid man-generator compilation warnings.
make.tmpl: Support per-object DEFS.
man: Preserve template variables in pre-generated pages.
man: Build man-generator in tools dir.
man-generator: Remove unused variable.
lvcreate: continue to accept --thinpool with -L and -V but not -T
tools: Show configuration command line in lvm version.
tools: Reinstate lvm script processing.
make: Don't hard-code SHELL as /bin/sh
man: pre-generated files weren't committed
pre-release
Bryn M. Reeves (66):
libdm-stats: fix dm_stats_delete_region() performance
libdm-stats: clear dms->regions in _stats_regions_destroy()
libdm-stats: clear dms->groups in _stats_groups_destroy()
libdm: use correct region_id when cleaning up a failed filemap
libdm: split off internal _stats_delete_region()
libdm: fix performance of failed filemap cleanup
doc: add filemap creation fixes to WHATS_NEW_DM
libdm: use a private pool for filemap extent table
libdm: fix filemap cleanup loop condition
scripts: add lvmdump.sh to .gitignore
scripts: add systemd unit files to .gitignore
lvmdbusd: add path.py to .gitignore
test: add results/ to .gitignore
libdm: fix _stats_get_extents_for_file()
libdm: fix region overlap tests
libdm: fix segfault with invalid group descriptor
libdm: check for non-existent region_id values in groups
libdm: fix dm_stats_foreach_group() macro
libdm: break up _stats_get_extents_for_file()
libdm: fix start of file detection in _stats_map_extents()
configure: check for __builtin_clz()
util: add clz() and use __builtin_clz() if available
libdm: add dm_bit_get_last()/dm_bit_get_prev()
libdm: use dm_bit_get_last() in _stats_group_tag_fill()
libdm: add min_num_bits to dm_bitset_parse_list()
libdm: ensure first extent is always counted
libdm: use destination size as limit in dm_bit_copy()
man: fix 'dmstats <command>' formatting in dmstats.8.in
man: fix 'dmstats list' option formatting in dmstats.8.in
man: fix 'dmstats create' formatting in dmstats.8.in
dmstats: improve tool help output and option coverage
dmstats: use better interval estimate for usleep() timer
dmstats: separate TIMERFD and useleep() exit conditions
dmstats: fix interval number reporting with --count=0
dmstats: fix TIMERFD _timer_running() test
dmstats: don't declare _start_timestamp if HAVE_SYS_TIMERFD_H
libdm: clear region table in dm_stats_list()
libdm: add dm_stats_bind_from_fd()
man: fix name of 'write_time' field in dmstats.8.in
libdm: fix stats comment formatting in libdevmapper.h
libdm: test for DM_STATS_GROUP_NOT_PRESENT in _stats_group_id_present
libdm: add dm_stats_update_regions_from_fd()
dmstats: allow --filemap groups to be updated
man: add update_filemap to dmstats.8.in
man: mention dmstats in lvmsar and lvmsadc man pages
libdm: remove unnecessary backtrace in _stats_group_id_present()
daemons: add dmfilemapd
man: add dmfilemapd.8
dmstats: start dmfilemapd when creating or updating file maps
man: add dmfilemapd options to dmstats.8.in
man: add FILE MAPPING section to dmstats.8.in
configure: make --enable-dmfilemapd require fiemap.h
libdm: don't nest FIEMAP and DMFILEMAPD ifdefs
dmfilemapd: always set the program_id after listing regions
libdm: move initialisation of group_id in _aggregate_histogram()
dmsetup: always initialise 'mode' in filemap create functions
tools: add generated headers to .gitignore
man: add man-generator and build artefacts to .gitignore
libdm: dm_stats_start_filemapd() mode can never be < 0
libdm: make dm_stats_destroy(NULL) a no-op
dmsetup: always initialise 'dms' in _stats_update_file()
libdm: use correct value for NR_FILEMAPD_ARGS
daemons: cleanup dmeventd references from dmfilemapd/Makefile.in
daemons: use DEFAULT_PROC_DIR in dmfilemapd
dmfilemapd: return error for wrong number of args
dmfilemapd: ensure path argument is absolute
Christian Brauner (1):
lvmetad: fix segfault on i386
David Teigland (150):
lvmetad: fix segfault in daemon_reply_simple
lvmlockd: test mode doesn't work
commands: new method for defining commands
process_each_lv: add check_single_lv function
lvchange: make use of command definitions
lvconvert: repair and replace: use command definitions
lvconvert: remove unused calls for repair and replace
lvconvert: snapshot: use command definitions
lvconvert: add startpoll command using command def
lvconvert: use command defs for thin/cache/pool creation
toollib: find VG name in option values when needed
lvconvert: use command defs for mergemirrors
tests: use swapmetadata
lvconvert: use command defs for raid/mirror types
lvconvert: remove unused code
args: use uint32 arg for maxphysicalvolumes
args: add man page descriptions
man lvm: remove options
fix lvmcmdline warning
help: accept positional args
lvm shell: clear argv for each command
command.h comment tidying
command struct: remove command name refs
ccmd: split into multiple files
man: lvmthin updates
args: split is a synonym for splitcache
lvconvert: remove unused code
commands: move command def parsing into lvm binary
commands: recognize raid variations
vgchange: fix uint32 parsing of logicalvolume arg
lvconvert: add command to change region size of a raid LV
lvconvert: remove code for changing region size
args: use arg parsing function for region size
generate man pages
commands: suggest --help if command doesn't match
lvconvert: enable previous syntax to swap metadata
WHATS_NEW: items from command definitions patch set
WHATS_NEW: changes for recent command defs
coverity cleanups for recent command defs commits
commands: improve help/usage output
commands: skip parsing command defs for other command names
man/help: use separator in units list
man-generator: fix compiler warnings
coverity cleanups
Revert "lvconvert: enable previous syntax to swap metadata"
lvcreate: LVM_VG_NAME applies even without name arg
lvconvert: handle old arg quirk when combining snapshot
toollib: skip dev prefix from lv names
lvcreate: fix LVM_VG_NAME detection
commands: use existing vg/lv name parsing functions
commands: use dm wrappers for malloc/free
help/man: remove commas from list of options
commands: avoid factoring options until needed
commands: initialize commands[] once
commands: track errors in command def parsing
lvconvert: include swap behavior in generic pool syntax
args: in yes option mention qq for auto-no
man/help: show recently added r|R unit
man lvmconfig: refer to typeconfig option name
man: mention force behavior in various places
man pvmove: move details to notes
man: print space before built-in name
man/help: fix description of r|R unit
lvresize/lvextend: accept type option
man: add variables descriptions
man lvcreate: add special option notes at the start
man vgchange/lvchange: note special option usage
man/help: print general raid type as "raid"
man: use .HP instead of .TP
man: expand type option description
man lvm: update see also list
commands: adjust default help output
man: teaking output format
man/help: use separate common option lists
man/help: enclose each option in brackets
man: improve writemostly PV arg
man: indicate repeatable options
man/help: improve the style of units indicator
man/help: change syntax to UNIT
man: improve line breaks
man: use Size variable for a number with unit
lvconvert: fix handling args in combining snapshots
man: add ENVIRONMENT VARIABLES section
man: change option sorting in synopsis
commands: fixes for recent raid changes
commands: fix lvconvert raid rule
man: mention regionsize default is in lvm.conf
commands: remove lvconvert raid rule
man: fix syntax for PV ranges
commands: include lvconvert cache options as group
help: print full usage for lvm help all
commands: SECONDARY flag changes in cmd defs
man: put some commands into advanced usage section
commands: tweak some descriptions
man: fix typo
man lvchange: mention special activation vals
commands: adjust syntax error message
lvcreate/lvresize: the --size option accepts signed values
commands: handle including an optional opt multiple times
lvcreate: allow chunksize option when creating cache
man: change the synopsis option style
lvcreate: munge size value in help output
help: print info about special options and variables
help: show extents option for lvcreate
man lvcreate: remove the extents prefix
man lvcreate: show extents option
args: update select description
args: in cachemode option fix passthrough value
man/help: improve the PV range description
man/help: rework extents and size output
help: show short opt with long opt
man/help: poolmetadatasize option can take relative value
man: move the full UNIT description
man lvextend: mention segment type
man/help: improve stripes option wording
help: avoid end notes repetition in lvm help all
commands: combine duplicate arrays for opt and val
commands: combine duplicate arrays for lv types and props
commands: clean up and unify signed option value handling
tests: add test for relative option values
man: add more references to topical man pages
man/help: use order of required options from cmd def
help: fix missing required option
lvresize: poolmetadatasize cannot use minus
tests relative-sign-options.sh: skip without thin
tests relative-sign-options: only skip thin part
man lvmraid: remove fixmes, use consistent example name
commands: coverity fixes
commands: fix missing skip condition in coverity fixes
commands: fix lvconvert options for mirrors/stripes
commands: use correct relative signs with mirror option
commands: fix lvcreate mirrors/stripes options
commands: fix command matching for unused options
man clvmd: mention attempt for deactivate
args: clarify repeated --options lists
man: remove redundant line breaks
man: remove more unnecessary line breaks
man: remove blank lines
man: remove blank lines in included files
help: fix missing newlines
man: remove blank lines from end/des files
lvconvert: disable lvmetad for repair
WHATS_NEW: disable lvmetad for repair
lvmetad: fix bug in snprintf of disable reason
help: align option list in pv/lv/vgchange cases
man: advise against mirrored mirror log
Revert "lvcreate: continue to accept --thinpool with -L and -V but not -T"
commands: fix combined thin pool and vol create defs
commands: fix commands with run with path basename
commands: improve error message for unknown command
Heinz Mauelshagen (121):
lvchange: allow a transiently failed RaidLV to be refreshed
lvchange: allow a transiently failed RaidLV to be refreshed
lvconvert: add segtype raid5_n and conversions to/from it
lvconvert: add segtype raid5_n and conversions to/from it
WHATS_NEW: New segment type raid5_n
lvconvert: add segtype raid6_n_6 and conversions to/from it
WHATS_NEW: New segment type raid6_n_6
lvconvert: add segtypes raid6_{ls,rs,la,ra}_6 and conversions to/from it
lvconvert: add support to change RAID region size
lvconvert: add segtype raid5_n and conversions to/from it (cleanup)
lvconvert: add support to change RAID region size (fixup)
lvconvert: avoid setting segment flag
lvconvert: add support to change RAID region size (fixup)
lvreduce/lvresize: add ability to reduce the size of a RaidLV
lvconvert: add conversion from/to raid10
test: lvresize-raid.sh missing conditional
lvconvert: add region size checks
lvconvert: remove test code
lvchange: reject writemostly/writebehind on raid1 during resync
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
raid: typo
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert: add infrastructure for RaidLV reshaping support
lvconvert/lvcreate: raise maximum number of raid images
lvconvert: allow regionsize on upconvert from linear
lvconvert: add missed new test scripts for reshaping
WHATS_NEW: add entry
lvconvert: fix 2 issues identified in intesting
lvconvert: impose region size constraints
lvconvert: limit libdm to maximum of 64 RAID devices
lvconvert: libdm RAID API compatibility versioning
lvconvert: adjust reshaping check to target version
lvconvert: adjust mininum region size check
lvconvert: add missing reshape_len initialization
raid: rework _raid_target_present()
lvconvert: add new reporting fields for reshaping
lvconvert: libdm RAID API compatibility versioning; remove new function
lvconvert: add new reporting fields for reshaping
lvs: enhance stripes and region size field descriptions
man: cover reshaping
dev_manager: remove reshape message
man: fix term in lvmraid(7)
report: raid enhancements for --select
test: fix typo
man lvs: describe new 'R' volume health character
test: add delay to lvchange-raid1-writemostly.sh
report: display proper LV size for reshapable RaidLVs
lvconvert: adjust --stripes on raid10 convert
raid: cleanup _lv_set_image_lvs_start_les()
raid: fix raid LV resizing
report: correct lv_size for 2-legged raid5
raid: define seg->extents_copied
metadata: comments
report: fix segfault
test: Copyright
lvconvert: fix --repair after vgreduce
raid: fix function description
test: "lvconvert --repair" after vgreduce
lvconvert: prompt when converting raid1 to linear
lvconvert: prompt when splitting off a tracked LV of a 2-legged raid1 LV
test: raid1 down convert to linear / split and track tests
lvconvert: remove superfluous compile time conditonal code
test: raid1 down convert to linear
test: raid1 down convert to linear
lvconvert: prompt when splitting off LV of a 2-legged raid1 LV
dmeventd: (workaround) fix mirror DSO to work with lvmetad
raid: add missing lv_merge_segments() call
man: add more examples to lvmraid(7)
lvconvert: add raid1 <-> raid4 conversion
raid: fix compile time warning
lvcreate: fix "striped" limit
raid: enhance lv_raid_convert() header relative to reshaping
raid: fix compile time warning
man: add more info/examples to lvmraid(7)
report: fix lv_size with reshape space allocated
man: fix typo
raid: check more cautious on region size changes
dmeventd: reintroduce fix mirror DSO to work with lvmetad
dmeventd: adjust mirror/raid DSOs to new repair design
test: adjust to proper dm-raid target version
lvconvert: ensure upconversion restrictions
WHATS_NEW: adjust mirror+raid DSOs to lvconvert --repair
WHATS_NEW: ensure raid6 upconversion restrictions
man: lvmraid(7) clarifications
raid: conditionally reject convert to striped/raid0*
raid: name variables consistently
raid: cleanup
raid: cleanup; remove unused function
WHATS_NEW: conditionally reject raid convert
raid: conditionally reject convert to striped/raid; fix
raid: check target version for shrink support
raid: adjust to misordered raid table line output
WHATS_NEW: adjust to misordered raid parameters
metadata: cleanup flags definition to be consistent
Revert "raid: adjust to misordered raid table line output"
raid: adjust to misordered raid table line output
libdm: typo
man: escape all double '-'
man: revert erouneous '-' escapes in Makefine.in
man: escape all single '-'
man-generator: emit escaped '-'
man-generator: cleanup escape '-'
WHATS_NEW: man-generator escape '-'
man/help: avoid escaping of '-' with --help
man-generator: fix buffer length calculation
man-generator: more escaped '-'
man: a few more missed '-' to escape
lvchange: reject setting all raid1 images to writemostly
man-generator/man/help: simplify hyphen escaping
man: enhance man postprocessing regexp
man: fix / typo
Jonathan Brassow (1):
man: Minor clarifications and grammar corrections for lvmraid.7
Marian Csontos (8):
man: Handle empty MAN8GENERATED
spec: lvmchange is built-in command
man: Make dmfilemapd.8 optional
spec: Profiles are not %config(noreplace)
spec: Update requirements for lockd
spec: Use %global instead of %define for constants
spec: Replace remaining %define by %global
spec: Disable sanlock only, keep lockd_dlm enabled
Mikulas Patocka (1):
build: fix x32 arch
Ondrej Kozina (1):
dmsetup: do not suppress kernel key descriptions in tables
Peter Rajnoha (2):
report: order fields by type for field defintions in columns.h
blkdeactivate: also unmount mount point on top of MD device if using blkdeactivate -u
Tony Asleson (34):
lvmdbusd: Remove un-needed main thread execution
lvmdbusd: Use cfg.reload() instead of dbo.refresh
lvmdbusd: Use timeout_add instead
lvmdbustest: Print messages if timeout value > 10%
lvmdbusd: lv.py, remove duplicate code
lvmdbusd: vg.py, remove duplicate code
lvmdbusd: pv.py, remove duplicate code
lvmdbusd: lvmdb.py, remove duplicate code
lvmdbusd: manager.py, remove duplicate code
lvmdbusd: cmdhandler.py, remove duplicate code
lvmdbustest.py: Use _lookup function
lvmdbustest.py: Verify lookups work immediately after vg create
lvmdbustest.py: Re-name validation function
lvmdbustest.py: Validate PV device
lvmdbustest.py: Validate LV lookups
lvmdbustest.py: Remove un-used variable
lvmdbusd: Add FlightRecorderDump
lvmdbusd: Use work queue for queries too
lvmdbusd: Disable notify_dbus for service command use
lvmdbusd: Correct src doc for Manager.UseLvmShell
lvmdbustest.py: Remove un-used `import time`
lvmdbustest.py: Add unit test for external VG create
lvmdbustest.py: Add unit test for external LV create
lvmdbustest.py: Add unit test for external PV create
lvmdbusd: Add debug message to lvm ID lookup
lvmdbusd: Correct utils.add_no_notify doc
lvmdbusd: Simplify add_no_notify for loop
lvmdbustest.py: Correct call_lvm doc
lvmdbustest.py: Correct debug & assert order
lvmdbusd: Only disable notify_dbus after getting external event
lvmdbusd: Call add_no_notify for *move commands
lvmdbusd: Limit state refreshes for udev events
lvmdbusd: Update state during pv move
lvmdbusd: Rename ee to got_external_event
Zdenek Kabelac (231):
snapshot: reporting uses statusinfo
striped: implement compatible target name
lv: always check status type
activation: status check switch to warn
activation: improve error handling for status reading
cleanup: swap test order
activation: lv_info_with_seg_status unify status selection
activation: lv_info_with_seg_status API change
activation: optimize away lv_has_target_type
dev_manager: extend setup_task
dev_manager: use setup_task_run for mknod
cleanup: simplify code
cleanup: indent and messsages updates
cleanup: simplier code
debug: add missing backtrace
tests: check we recognize broken table entry
tests: apostrof
tests: more snaps
cleanup: messages in raid
cleanup: easier code for raid plugin
cleanup: zero baton in struct initilizer
tests: fix missing exclusive activation
mirror: add prepare_mirror_log
raid: fix raid1 to mirror conversion
raid: fix delete on clustered vg
raid: optimize clearing of lvs
raid: avoid _ at end of name of extracted metadata LV
pvmove: fix activation order
libdm: validate vsnprintf
cleanup: reuse existing function
cleanup: use NAME_LEN stack buffer
debug: missing stack traces
cov: add internal error for impossible code path
cov: declaration matching
tests: slower device
tests: test should clean devices it has created
tests: remove some leaking device
tests: remove unwanted exit
tests: track leaked devices in tests
cov: use unsigned for single bit values
tests: no left devices check for skipped test
tests: use prepare_vg more often
tests: update seg_size_pe
raid: improve table reload sequence
segtype: check for seg type instead of status
raid: avoid manipulation of segment status
cleanup: allocate NAME_LEN size for lv name
cleanup: more lv_is_ usage
cleanup: log message updates
cleanup: remove wrapping function
configure: just move new macro to right file
raid: activation with list
raid: split preserves local exlusive activation
cleanup: use exiting function
raid: fix activation of tracked image
lvs: show status for layer
thin: reload external origin with last thin
validation: check external property is matching
cache: improve wait for cache clear
cache: prepare status checking for layer
lv: fix lock holder for external origin
libdm: drop callback on revert path
cache: improve activation with -real
lvconvert: support cache to external origin conversion
backup: show warning once per command
thin: add comment with future extension
activate: further _info API refinement
debug: add debug message showing new lv
tests: using cached LV for external origin
lvconvert: fix shown lv name for snapshot split
cache: support cached origin for snapshot
tests: usage of cached volume for snapshot
validation: rework segment validation
lv: more exact check for merging origin
snapshot: validate merge has started
report: show proper info for merging origin
thin: refresh status when error processing fails
tests: workaround failure on fc23
tests: add device holding function
tests: use hold_device_open
validation: temporarily let pass linear with chunk_size
tests: update test
lvmcmdline: support uint32
vgchange: max_pv limited to uint32
cache: add missing udev wait
cleanup: reuse existing code
cleanup: defines
cleanup: use macros
cleanup: more use of lvseg_name
cache: resize is still unsupported
debug: show proper error message for layer mismatch
report: report merged state for inactive LV
tests: drop FIXME
tests: slow down devs for raid more
mirror: relax internal error for a while
vgchange: also -l is uint32
raid: postpone archiving until metadata are changed
tests: make test more race immune
tests: ignore racy test failure
libdm: add human R|readable units
cleanup: update config doc
config: new option dmeventd/thin_command
lvmcmd2lib: support new command
dmeventd_thin: handling of internal command
dmeventd_thin: switch to struct percent
dmeventd_thin: better warning logic
dmeventd_thin: add run_command
dmeventd_thin: add wait_pid
dmeventd_thin: init command
dmeventd_thin: SIGCHLD handler
dmeventd_thin: rework failure handling
dmeventd_thin: drop umounting on error path
dmeventd_thin: new logic for calling commands
dmeventd_thin: enable support for external command
dmeventd_thin: internal command without lvm prefix
tests: enusure units in TiB
tests: implement umount in script
tests: properly quote heredoc
tests: umount when above 95
clean: move code to lib part
dmeventd_thind: set LVM_RUN_BY_DMEVENTD
toolcontext: action for LVM_RUN_BY_DMEVENTD env var
clvmd: add mutex protection for cpg_ call
cleanup: rename to use track_ prefix
comment: update
debug: add space before uuid
cleanup: hide gcc warning
tests: drop /tmp polution
tests: drop zeroing
libdm: avoid resume if preloaded device is smaller
mem: add extra mem pages for pthread stack
cleanup: drop double const specifier
coverity: remove unneeded header files
coverity: fix mem leak on error path in dm stats
coverity: fix double free
dmeventd: limit thin_command usage
config: update doc
cleanup: use proper printf specifier
cleanup: fix warning about shadowed declaration
cleanup: use fall through
cleanup: use matching const type
cleanup: use matching signed number comparation
tests: fix dev path
tests: update
dmeventd: unify error handling
activation: trust 'pending' status reported by dmeventd
configure: update
libdm: move code for _stats_resize_group
makefiles: use configure vars
makefiles: missing deps
commands: cheap optimisation for parser
commands: use dm_strncpy
commands: optimize binary search
cleanup: add comment for longer macro
cleanup: use strchr for 1-byte lookup
cleanup: unused uninit vars
commands: some more dm wrappers
commands: compensate --uuid
lvresize: fix regressiong on accepting args
commands: fix memleak
man: dmsetup update
man: dmstats update
man: thin dmeventd plugin update
cleanup: use const
cleanup: use multiline string
cleanup: same sign for compare
cleanup: simplier code
tests: update lvresize testing
tests: use exclusive activation for mirror
tests: paralelize creation and teardown
tests: make sure which PV is used for allocation
tests: ensure first call is version test
tests: some debug code
WHATS_NEW: add entry
snapshot: improve removal of active origin volume
tests: not slowing work with metadata
tests: support zero_dev
lvconvert: indent and code simplification
lvconvert: validate name before prompt
lvcreate: avoid rejecting --metadataprofile
cache: properly translate DM_THIN_DISCARDS
thin: getting default chunk_size from single place
pool: split chunk size validation
pool: drop duplicated sign test
thin: add new ZERO/DISCARDS_UNSELECTED
cache: use UNSELECTED enum
cache: set chunk_size as first param
lvconvert: replacing with goto
cache: get and set cache params
cache: improve support for profile for cache settings
cache: extend usability of cache_set_params
lvcreate: respecting profile settings
lvcreate: do not round cache volumes on cache chunks
lvconvert: share function for cache params
parser: allow --zero for pool conversion
parser: enable --metadataprofile for pool lvconvert
thin: single formula for estimation
pool: rework handling of passed args
cache: improve profile support for cache_set_policy
cleanup: minor cosmetics
tests: use LVM_SYSTEM_DIR for etc
libdm: passthrough checks traling space char
libdm: better code to enforce writethrough
libdm: support cache metadata2 feature flag
libdm: maintain binary interface for new FEATURE flag
cache: introduce allocation/cache_metadata_format
cache: LV supports cache segs with metadata format
cache: use DM_CACHE_FEATURE_METADATA2 with libdm
cache: validation for cache_metadata_format
cache: reporting cache metadata format
cache: allow to specify cachemetadataformat
cache: enable usage of --cachemetadataformat
cache: enhance lvdisplay for cache volumes
cleanup: easier code
cleanup: use FMTu
tests: cache profile and metadata2
mirror: avoid scanning
coverity: update models
thin: remove unneeed test for NULL
libdm: fix endless loop
coverity: make sure segtype pointer is valid
raid: use 64bit arithmetic
mirror: revert 76f6951c3e8f0933df9730a42e9c46f273d1da24
tests: add missing --yes
tests: better disk delay
lvm2app: correct internal API changes
properties: only thin-pool provides discards
tests: raise min dm cache version
tests: early detect leaking error dev
cleanup: use more common FMTd64 type
tests: update to better fit