]> sourceware.org Git - lvm2.git/log
lvm2.git
4 years agodmeventd: enhance time waiting loop
Zdenek Kabelac [Wed, 4 Mar 2020 14:56:09 +0000 (15:56 +0100)]
dmeventd: enhance time waiting loop

dmeventd is 'scanning' statuses in loop (most usually in 10sec
intervals) - and meanwhile it sleeps within:
pthread_cond_timedwait()

However this function call tends to wakeup sometimes a short amount of
time sooner - and our code still believe the 'right time' has not yet
arrived and basically for a moment 'busy-looped' on calling this
function - so for systems with 'clock_gettime()' present we obtain
time and we go 10ms to the future second - this avoids unneeded
repeated invocation of our time scheduling loop.

TODO: monitoring during 1 hour 'time-change'...

4 years agopvck: use dm_config_parse_without_dup_node_check
David Teigland [Wed, 4 Mar 2020 17:32:13 +0000 (11:32 -0600)]
pvck: use dm_config_parse_without_dup_node_check

instead of dm_config_parse.  Some strange case could
cause dm_config_parse to print duplicate warnings about
all the metadata fileds.

4 years agotests: reduce sizes in pvck-dump and improve checks
David Teigland [Wed, 4 Mar 2020 17:30:50 +0000 (11:30 -0600)]
tests: reduce sizes in pvck-dump and improve checks

Smaller devs can be used so tests can be run on small vms.
Improve checks.

4 years agotests: pvck dump from larger metadata areas
David Teigland [Tue, 3 Mar 2020 19:44:59 +0000 (13:44 -0600)]
tests: pvck dump from larger metadata areas

4 years agopvck: allow dump from file
David Teigland [Mon, 2 Mar 2020 22:26:44 +0000 (16:26 -0600)]
pvck: allow dump from file

4 years agopvck: fix reading large mda1
David Teigland [Mon, 2 Mar 2020 21:22:39 +0000 (15:22 -0600)]
pvck: fix reading large mda1

When mda_size is larger than io_memory_size, reading
the entire mda fails unless the previous read of the
label has been invalidated.

4 years agopvck: improve mda_offset mda_size choices
David Teigland [Fri, 28 Feb 2020 18:00:23 +0000 (12:00 -0600)]
pvck: improve mda_offset mda_size choices

Attempt to calculate an offset or size if one only
value was specified in the settings.

Use header values when available.

4 years agopvck: print longer command description
David Teigland [Fri, 28 Feb 2020 17:24:28 +0000 (11:24 -0600)]
pvck: print longer command description

4 years agopvck: ensure text lines are terminated
David Teigland [Fri, 28 Feb 2020 17:20:27 +0000 (11:20 -0600)]
pvck: ensure text lines are terminated

4 years agohints: free hint list in error exit path
David Teigland [Tue, 3 Mar 2020 18:25:34 +0000 (12:25 -0600)]
hints: free hint list in error exit path

4 years agoman: lvmcache raid1 references
Jonathan Brassow [Thu, 27 Feb 2020 17:33:55 +0000 (11:33 -0600)]
man: lvmcache raid1 references

4 years agotests: validate vdo slab_size
Zdenek Kabelac [Wed, 26 Feb 2020 12:26:46 +0000 (13:26 +0100)]
tests: validate vdo slab_size

New vdoformat can print this size - so check we pass proper bit count
matching preset value.

4 years agovdo: fix slab size bits calculation
Zdenek Kabelac [Fri, 21 Feb 2020 10:59:29 +0000 (11:59 +0100)]
vdo: fix slab size bits calculation

When formating VDO volume, the calculated amound of bits
for 'vdoformat --slab-bits' parameter was shifted by 2 bits
(calculated size was making 2MiB vdo_slab_size_mb value appear like if
user would be specifying only 512KiB)

Fixed by properly converting internal size_mb value to KiB.

4 years agowritecache: check watermark value
David Teigland [Tue, 25 Feb 2020 16:34:30 +0000 (10:34 -0600)]
writecache: check watermark value

4 years agowritecache: allow removing wcorig lv
David Teigland [Fri, 21 Feb 2020 18:40:42 +0000 (12:40 -0600)]
writecache: allow removing wcorig lv

like removing corig

4 years agowritecache: fix watermark error message
David Teigland [Fri, 21 Feb 2020 14:13:32 +0000 (08:13 -0600)]
writecache: fix watermark error message

4 years agowritecache: working real dm uuid suffix for wcorig lv
David Teigland [Wed, 19 Feb 2020 19:30:02 +0000 (13:30 -0600)]
writecache: working real dm uuid suffix for wcorig lv

4 years agowritecache: drop real dm suffix
David Teigland [Mon, 17 Feb 2020 18:14:13 +0000 (12:14 -0600)]
writecache: drop real dm suffix

fixes the problem of adding writecache to an active LV

4 years agothin: don't use writecache for poolmetadata
David Teigland [Thu, 13 Feb 2020 23:22:37 +0000 (17:22 -0600)]
thin: don't use writecache for poolmetadata

4 years agowritecache: check if cachevol is writable
David Teigland [Tue, 11 Feb 2020 18:59:39 +0000 (12:59 -0600)]
writecache: check if cachevol is writable

before trying to initialize it (since wipe_lv
does not return an error if it fails to write.)

4 years agocachevol: stop dm errors with uncaching cache with cachevol
Zdenek Kabelac [Tue, 11 Feb 2020 16:11:39 +0000 (17:11 +0100)]
cachevol: stop dm errors with uncaching cache with cachevol

Fix the anoying kernel message reported:
device-mapper: cache: 253:2: metadata operation 'dm_cache_commit' failed: error = -5
which has been reported while cachevol has been removed.
Happened via confusing variable - so switch the variable to commonly user '_size'
which presents a value in sector units and avoid 'scaling' this as extent length
by vg extent size when placing 'error' target on removal path.

Patch shouldn't have impact on actual users data, since at this moment
of removal all date should have been already flushed to origin device.

m

4 years agopost-release
Marian Csontos [Tue, 11 Feb 2020 09:53:01 +0000 (10:53 +0100)]
post-release

4 years agopre-release v2_03_08
Marian Csontos [Tue, 11 Feb 2020 09:50:50 +0000 (10:50 +0100)]
pre-release

4 years agovdo: fix vdoformat when -V is specified
Zdenek Kabelac [Mon, 10 Feb 2020 14:41:30 +0000 (15:41 +0100)]
vdo: fix vdoformat when -V is specified

The previous patch improved read of pipe when lvm2 was looking
for default logical size, but we clearly must read pipe also
for -V case, when the logical size is already defined.

4 years agowritecache: skip zeroing in test mode
David Teigland [Fri, 7 Feb 2020 16:32:10 +0000 (10:32 -0600)]
writecache: skip zeroing in test mode

4 years agowritecache: check for invalid cachevol
David Teigland [Fri, 7 Feb 2020 16:26:59 +0000 (10:26 -0600)]
writecache: check for invalid cachevol

4 years agowritecache: fix return value
David Teigland [Fri, 7 Feb 2020 16:21:07 +0000 (10:21 -0600)]
writecache: fix return value

4 years agoraid: better place for blocking reshapes
Zdenek Kabelac [Fri, 7 Feb 2020 14:14:05 +0000 (15:14 +0100)]
raid: better place for blocking reshapes

Still the place can be better to block only particular reshape
operations which ATM cause kernel problems.

We check if the new number of images is higher - and prevent to take
conversion if the volume is in use (i.e. thin-pool's data LV).

4 years agowritecache: prevent snapshots
David Teigland [Thu, 6 Feb 2020 17:27:33 +0000 (11:27 -0600)]
writecache: prevent snapshots

there appear to be problems with taking a snapshot
of an LV with a writecache, so block it until that
is understood or fixed.

4 years agowritecache: fix splitcache when origin is raid
David Teigland [Tue, 4 Feb 2020 22:05:56 +0000 (16:05 -0600)]
writecache: fix splitcache when origin is raid

4 years agoWHATS_NEW: update
Zdenek Kabelac [Tue, 4 Feb 2020 16:21:13 +0000 (17:21 +0100)]
WHATS_NEW: update

4 years agogenerate: remake
Zdenek Kabelac [Tue, 4 Feb 2020 14:40:33 +0000 (15:40 +0100)]
generate: remake

Regen man page.

4 years agolv_manip: add extra check for existin origin_lv
Zdenek Kabelac [Thu, 30 Jan 2020 14:54:19 +0000 (15:54 +0100)]
lv_manip: add extra check for existin origin_lv

clang: it's supposedly impossible path to hit, as we should always
have origin_lv defined when running this path, but adding protection
isn't a big issue to make this obvious to analyzer.

4 years agoraid: add internal error for no segment
Zdenek Kabelac [Thu, 30 Jan 2020 14:46:05 +0000 (15:46 +0100)]
raid: add internal error for no segment

clang: capture internal error when data_seg would not be defined.
(invalid LV with no areas)

4 years agolv_manip: add error handling for _reserve_area
Zdenek Kabelac [Thu, 30 Jan 2020 14:23:47 +0000 (15:23 +0100)]
lv_manip: add error handling for _reserve_area

Since _reserve_area() may fail due to error allocation failure,
add support to report this already reported failure upward.

FIXME: it's log_error() without causing direct command failure.

4 years agocommand: validate reporting of previous argument
Zdenek Kabelac [Thu, 30 Jan 2020 14:08:30 +0000 (15:08 +0100)]
command: validate reporting of previous argument

When reporting parsing error, report 'previous' argument
only when there is one.

4 years agodmeventd: nicer error path for reading pipe
Zdenek Kabelac [Wed, 29 Jan 2020 15:41:28 +0000 (16:41 +0100)]
dmeventd: nicer error path for reading pipe

When _daemon_read()/_client_read() fails during the read,
ensure memory allocated withing function is also release here
(so caller does not need to care). Also improve code readbility a bit
a for same functionality use more similar code.

4 years agolvmlockctl: use inline initilizers
Zdenek Kabelac [Wed, 29 Jan 2020 14:47:40 +0000 (15:47 +0100)]
lvmlockctl: use inline initilizers

clang: ensure r_name[] is in all possible paths defined.

4 years agolvmlockctl: ensure result value is always defined
Zdenek Kabelac [Tue, 4 Feb 2020 14:33:36 +0000 (15:33 +0100)]
lvmlockctl: ensure result value is always defined

Ensure passed pointer gets predefined value (instead of random stack
value).

4 years agolvmlockd: move eval of ENOENT
Zdenek Kabelac [Thu, 30 Jan 2020 11:52:19 +0000 (12:52 +0100)]
lvmlockd: move eval of ENOENT

To avoid logging 'errors' for no real error state (ENOENT),
move this evaluation upward in the code.

4 years agocov: check error code from mutex init
Zdenek Kabelac [Thu, 30 Jan 2020 11:48:40 +0000 (12:48 +0100)]
cov: check error code from mutex init

4 years agocov: widen before calculating min_chunk_size
Zdenek Kabelac [Thu, 30 Jan 2020 11:43:53 +0000 (12:43 +0100)]
cov: widen before calculating min_chunk_size

Although we expect min_chunk_size to be 32bit value, for
large size of caches it might be useful to do calcs 64bit.
So to avoid doing shift as signed 32bit - use unsigned 64bit
from the start.

4 years agocov: unused header file removal
Zdenek Kabelac [Thu, 30 Jan 2020 10:03:27 +0000 (11:03 +0100)]
cov: unused header file removal

cov: unused header removed
Also ensure library header file with config settings goes first.
Move inclusion of format-text.h into layout.h

4 years agobcache: fix memleak on error path
Zdenek Kabelac [Wed, 29 Jan 2020 16:31:17 +0000 (17:31 +0100)]
bcache: fix memleak on error path

clang: free io on error path.

4 years agopvck: avoid memleak of vgname
Zdenek Kabelac [Wed, 29 Jan 2020 15:55:49 +0000 (16:55 +0100)]
pvck: avoid memleak of vgname

clang: no vgname buffer leak.

4 years agodmsetup: no memleak on failed realocation
Zdenek Kabelac [Wed, 29 Jan 2020 15:30:16 +0000 (16:30 +0100)]
dmsetup: no memleak on failed realocation

clang: keep old buf pointer for release on failing realloc() codepath.

4 years agotests: compatible with older kernels
Zdenek Kabelac [Wed, 29 Jan 2020 14:50:57 +0000 (15:50 +0100)]
tests: compatible with older kernels

Older kernels just show syscall stacktrace.

4 years agotests: move vgsplit writecache
David Teigland [Tue, 4 Feb 2020 16:20:19 +0000 (10:20 -0600)]
tests: move vgsplit writecache

to a different file

4 years agopvmove: prevent moving writecache device
David Teigland [Mon, 3 Feb 2020 21:59:12 +0000 (15:59 -0600)]
pvmove: prevent moving writecache device

4 years agovgsplit: handle cachevol
David Teigland [Mon, 3 Feb 2020 21:33:58 +0000 (15:33 -0600)]
vgsplit: handle cachevol

attached to a cache or writecache LV.
Ensure PVs in cachevol are moved with the main LV.

4 years agowritecache: reject invalid high/low watermark setting
David Teigland [Mon, 3 Feb 2020 17:33:30 +0000 (11:33 -0600)]
writecache: reject invalid high/low watermark setting

4 years agoman: lvmcache writecache watermark percent
David Teigland [Mon, 3 Feb 2020 17:21:24 +0000 (11:21 -0600)]
man: lvmcache writecache watermark percent

4 years agowritecache: report status fields
David Teigland [Fri, 31 Jan 2020 17:52:49 +0000 (11:52 -0600)]
writecache: report status fields

reporting fields (-o) directly from kernel:
writecache_total_blocks
writecache_free_blocks
writecache_writeback_blocks
writecache_error

The data_percent field shows used cache blocks / total cache blocks.

4 years agoman: updates to lvmcache
David Teigland [Thu, 30 Jan 2020 20:09:21 +0000 (14:09 -0600)]
man: updates to lvmcache

4 years agolvmlockd: use ret value in query function
David Teigland [Wed, 29 Jan 2020 16:37:28 +0000 (10:37 -0600)]
lvmlockd: use ret value in query function

4 years agoWHATS_NEW: update
Zdenek Kabelac [Thu, 23 Jan 2020 09:25:35 +0000 (10:25 +0100)]
WHATS_NEW: update

4 years agotests: add corruption write on PV test
Zdenek Kabelac [Mon, 20 Jan 2020 13:16:13 +0000 (14:16 +0100)]
tests: add corruption write on PV test

Test a case where PV is readable, but fails on write updating.
Check the failure is reported only for a single PV.

4 years agotests: writeerror_dev
Zdenek Kabelac [Mon, 20 Jan 2020 12:15:46 +0000 (13:15 +0100)]
tests: writeerror_dev

Intruduce aux function for easy simulation of disk areas,
that are 'normally' readable, but will fail on write.

4 years agovdo: adapt for multi line vdo_format output
Zdenek Kabelac [Wed, 22 Jan 2020 11:57:23 +0000 (12:57 +0100)]
vdo: adapt for multi line vdo_format output

Do not close pipeline after 1st. line parsed from vdo_format.
Also reprint the output for a user so new messages from vdo_format
can be seen by users.

4 years agoraid: more limitted prohibition of stacked raid usage
Zdenek Kabelac [Thu, 16 Jan 2020 15:43:55 +0000 (16:43 +0100)]
raid: more limitted prohibition of stacked raid usage

We actually need to prohibit only reshaping cases which are
running over multiple commands.

4 years agoman: pvck dump description improvements
David Teigland [Wed, 22 Jan 2020 21:01:00 +0000 (15:01 -0600)]
man: pvck dump description improvements

4 years agoadd suggestion message for mirror LVs
Heming Zhao [Wed, 15 Jan 2020 02:44:30 +0000 (02:44 +0000)]
add suggestion message for mirror LVs

Currently the error messages are not clear. This very easy to
guide user to execute "--removemissing --force", it is dangerous
and will make the LVs to be destroied.

Signed-off-by: Zhao Heming <heming.zhao@suse.com>
4 years agofix corosync.conf: no interface error
Heming Zhao [Wed, 15 Jan 2020 02:17:47 +0000 (02:17 +0000)]
fix corosync.conf: no interface error

systemctl status corosync (version: 2.4.5) report error:
  parse error in config: No interfaces defined

Signed-off-by: Zhao Heming <heming.zhao@suse.com>
4 years agoWHATS_NEW: update
Zdenek Kabelac [Mon, 13 Jan 2020 16:37:29 +0000 (17:37 +0100)]
WHATS_NEW: update

4 years agotests: update for readahead
Zdenek Kabelac [Thu, 12 Dec 2019 14:54:17 +0000 (15:54 +0100)]
tests: update for readahead

4 years agovdo: avoid running initialization of cache pool vars
Zdenek Kabelac [Mon, 6 Jan 2020 14:57:08 +0000 (15:57 +0100)]
vdo: avoid running initialization of cache pool vars

Since VDO is also pool, the old if() case missed to know about this,
and executed unnecesserily initialization of cache pool variables.
This was usually harmless when using 'smaller' sizes of VDO pools,
but for big VDO pool size, we were reporting senseless messages
about big cache chunk sizes.

4 years agoraid: disallow reshape of stacked LVs
Zdenek Kabelac [Mon, 13 Jan 2020 16:27:24 +0000 (17:27 +0100)]
raid: disallow reshape of stacked LVs

Until we resolve reshape for 'stacked' devices, we need to disable it.
So users can no longer reshape i.e. thin-pool data volumes, causing
ATM bad thin-pool problems.

4 years agoWHATS_NEW: VDO lvmdbusd adds
Tony Asleson [Thu, 9 Jan 2020 19:11:41 +0000 (13:11 -0600)]
WHATS_NEW: VDO lvmdbusd adds

4 years agolvmdbusd: Add function to convert LV into a VDO pool
Vojtech Trefny [Mon, 6 Jan 2020 10:28:41 +0000 (11:28 +0100)]
lvmdbusd: Add function to convert LV into a VDO pool

4 years agolvmdbusd: Add VDO enable/disable compress & dedup
Vojtech Trefny [Fri, 27 Dec 2019 14:29:15 +0000 (15:29 +0100)]
lvmdbusd: Add VDO enable/disable compress & dedup

Added methods to vdo pool interface to allow enabling and
disabling of VDO:
 * Compression
 * Deduplication

4 years agotests: Some lvmdbus tests require larger PVs
Marian Csontos [Wed, 18 Dec 2019 14:33:58 +0000 (15:33 +0100)]
tests: Some lvmdbus tests require larger PVs

4 years agotests: VDO detection in dbus tests dev-mcsontos-tests-201912
Marian Csontos [Mon, 16 Dec 2019 10:25:37 +0000 (11:25 +0100)]
tests: VDO detection in dbus tests

4 years agodrop warnings about missing pvs in foreign vgs
David Teigland [Wed, 11 Dec 2019 18:56:15 +0000 (12:56 -0600)]
drop warnings about missing pvs in foreign vgs

When a foreign VG is ignored, don't print warnings that
it is missing PVs.

4 years agopvck: show specific dump option values
David Teigland [Tue, 10 Dec 2019 17:06:45 +0000 (11:06 -0600)]
pvck: show specific dump option values

4 years agotests: improve secure test
Zdenek Kabelac [Tue, 10 Dec 2019 12:43:47 +0000 (13:43 +0100)]
tests: improve secure test

Validate we capture core while original task sleeps.

4 years agotest: fix missing waiting on udev
Zdenek Kabelac [Tue, 10 Dec 2019 12:40:56 +0000 (13:40 +0100)]
test: fix missing waiting on udev

After device creation we need to wait for a cookie so it's not forgotten
in the system.

4 years agotest: fail on device create
Zdenek Kabelac [Tue, 10 Dec 2019 12:30:55 +0000 (13:30 +0100)]
test: fail on device create

Correct validation of prepared device and fail if the device can't
be created.

4 years agotest: aux setup
Zdenek Kabelac [Tue, 10 Dec 2019 12:30:11 +0000 (13:30 +0100)]
test: aux setup

Avoid endless loop if there was no 'remove' progress.

4 years agoclenaup: simpler form
Zdenek Kabelac [Thu, 28 Nov 2019 14:15:10 +0000 (15:15 +0100)]
clenaup: simpler form

4 years agovg_read: use else for 3 case
Zdenek Kabelac [Thu, 28 Nov 2019 14:09:27 +0000 (15:09 +0100)]
vg_read: use else for 3 case

Make it visible we check for ==, >, <  of same var.

4 years agodebug: enhance messages
Zdenek Kabelac [Thu, 28 Nov 2019 12:19:44 +0000 (13:19 +0100)]
debug: enhance messages

Drop 'extra' stack trace where errors are already logged from function.
Add some missing dots in messages.

4 years agodebug: avoid to slashes in debug message
Zdenek Kabelac [Thu, 24 Oct 2019 14:36:01 +0000 (16:36 +0200)]
debug: avoid to slashes in debug message

4 years agodebug: print VG name in log messages for segment errors
Nikhil Kshirsagar [Tue, 26 Nov 2019 07:06:45 +0000 (12:36 +0530)]
debug: print VG name in log messages for segment errors

Signed-off-by: Nikhil Kshirsagar <nkshirsa@redhat.com>
4 years agolibdm: set maj:min while creating and reloading device
Zdenek Kabelac [Tue, 10 Dec 2019 12:28:44 +0000 (13:28 +0100)]
libdm: set maj:min while creating and reloading device

Add maj:min to the task structure for RELOAD - which is now
handled in _flatten() and will just skip passing device name.

4 years agolibdm: support device RELOAD with maj:min and devname set
Zdenek Kabelac [Tue, 10 Dec 2019 12:28:16 +0000 (13:28 +0100)]
libdm: support device RELOAD with maj:min and devname set

When devices are created - we were not giving meaning error messages
when the failure happened on 'reload' part of creation.

With this patch we are now able to report both name and major:minor.

Enhancment is most visible with 'crypto' devices,
which are using 'secure' memory erase bit.

4 years agoman pvck: describe settings
David Teigland [Fri, 6 Dec 2019 22:21:48 +0000 (16:21 -0600)]
man pvck: describe settings

4 years agoupdate option description for settings
David Teigland [Fri, 6 Dec 2019 22:21:26 +0000 (16:21 -0600)]
update option description for settings

4 years agopvck: deal with coverity warnings
David Teigland [Mon, 2 Dec 2019 17:16:02 +0000 (11:16 -0600)]
pvck: deal with coverity warnings

4 years agopost-release
Marian Csontos [Sat, 30 Nov 2019 13:46:56 +0000 (14:46 +0100)]
post-release

4 years agopre-release v2_03_07
Marian Csontos [Sat, 30 Nov 2019 13:45:51 +0000 (14:45 +0100)]
pre-release

4 years agobuild: make generate
Marian Csontos [Sat, 30 Nov 2019 13:24:22 +0000 (14:24 +0100)]
build: make generate

4 years agopvck: use zalloc in more places
David Teigland [Wed, 27 Nov 2019 17:17:15 +0000 (11:17 -0600)]
pvck: use zalloc in more places

4 years agopvck: repair headers and metadata
David Teigland [Tue, 29 Oct 2019 21:08:43 +0000 (16:08 -0500)]
pvck: repair headers and metadata

To write a new/repaired pv_header and label_header:

  pvck --repairtype pv_header --file <file> <device>

This uses the metadata input file to find the PV UUID,
device size, and data offset.

To write new/repaired metadata text and mda_header:

  pvck --repairtype metadata --file <file> <device>

This requires a good pv_header which points to one or two
metadata areas.  Any metadata areas referenced by the
pv_header are updated with the specified metadata and
a new mda_header. "--settings mda_num=1|2" can be used
to select one mda to repair.

To combine all header and metadata repairs:

  pvck --repair --file <file> <device>

It's best to use a raw metadata file as input, that was
extracted from another PV in the same VG (or from another
metadata area on the same PV.)  pvck will also accept a
metadata backup file, but that will produce metadata that
is not identical to other metadata copies on other PVs
and other areas.  So, when using a backup file, consider
using it to update metadata on all PVs/areas.

To get a raw metadata file to use for the repair, see
pvck --dump metadata|metadata_search.

List all instances of metadata from the metadata area:
  pvck --dump metadata_search <device>

Save one instance of metadata at the given offset to
the specified file (this file can be used for repair):

  pvck --dump metadata_search --file <file>
    --settings "metadata_offset=<off>" <device>

4 years agopvck: dump functions cleanup args and return vals
David Teigland [Mon, 4 Nov 2019 19:38:23 +0000 (13:38 -0600)]
pvck: dump functions cleanup args and return vals

4 years agopvck: dump show most recent metadata
David Teigland [Mon, 4 Nov 2019 17:32:23 +0000 (11:32 -0600)]
pvck: dump show most recent metadata

4 years agopvck: allow disk locations to be specified
David Teigland [Tue, 22 Oct 2019 18:28:45 +0000 (13:28 -0500)]
pvck: allow disk locations to be specified

using --settings:

mda_offset=<offset> mda_size=<size> can be used
in place of the offset/size that normally come
from headers.

metadata_offset=<offset> prints/saves one instance
of metadata text at the given offset, in
metadata_all or metadata_search.

4 years agopvck: move some arg processing
David Teigland [Mon, 21 Oct 2019 21:39:46 +0000 (16:39 -0500)]
pvck: move some arg processing

4 years agoscan: add simple scan to find a pvid
David Teigland [Wed, 30 Oct 2019 20:11:57 +0000 (15:11 -0500)]
scan: add simple scan to find a pvid

4 years agometadata: add vg_from_config_tree
David Teigland [Tue, 29 Oct 2019 19:31:37 +0000 (14:31 -0500)]
metadata: add vg_from_config_tree

Add cmd/fmt args to import functions so that
they can be used without the fid arg which.

4 years agoRevert "cov: use zalloc"
David Teigland [Thu, 14 Nov 2019 18:31:31 +0000 (12:31 -0600)]
Revert "cov: use zalloc"

This reverts commit 9af1d63b4d96facb0aa055a44c2f61eac01e13f2.

fixes folded into subsequent pvck commit

This page took 0.07058 seconds and 5 git commands to generate.