]> sourceware.org Git - lvm2.git/log
lvm2.git
14 years agoUse vg_mda_count() in vgdisplay.
Dave Wysochanski [Mon, 28 Jun 2010 20:32:21 +0000 (20:32 +0000)]
Use vg_mda_count() in vgdisplay.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoAdd mda location specific mda_copy constructor.
Dave Wysochanski [Mon, 28 Jun 2010 20:31:59 +0000 (20:31 +0000)]
Add mda location specific mda_copy constructor.

Because of the way mdas are handled internally, where a PV in a VG
has mdas on both info->mdas and vg->fid->metadata_areas list, we
need a location independent copy constructor for struct
metadata_area.  Break up the existing format-text specific copy
constructor into a format independent piece and a format dependent
piece.

This function is necessary to properly implement pv_set_mda_ignored().

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
Reviewed-by: Alasdair G Kergon <agk@redhat.com>
14 years agoAdd mda_locns_match() internal library function for mapping pv/device to VG mda.
Dave Wysochanski [Mon, 28 Jun 2010 20:31:38 +0000 (20:31 +0000)]
Add mda_locns_match() internal library function for mapping pv/device to VG mda.

A metadata_area is defined independent of the location.  One downside
is that there is no obvious mapping from a pv to an mda.  For a PV in
a VG, we need a way to start with a PV and end up with an MDA, if we
are to manage mdas starting with a device/pv.  This function provides
us a way to go down the list of PVs on a VG, and identify which ones
match a particular PV.

I'm not entirely happy with this approach, but it does fit into the
existing structures in a reasonable way.

An alternative solution might be to refactor the VG - PV interface such
that mdas are a list tied to a PV.  However, this seemed a bit tricky since
a PV does not come into existence until after the list of mdas is
constructed (see _vg_read() - we create a 'fid' and attach mdas to it,
then we go through them and attach pvs).

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
Reviewed-by: Alasdair G Kergon <agk@redhat.com>
14 years agoEnsure in-memory state matches on-disk state of mda ignore bit.
Dave Wysochanski [Mon, 28 Jun 2010 20:31:18 +0000 (20:31 +0000)]
Ensure in-memory state matches on-disk state of mda ignore bit.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoAllow raw_read_mda_header to be called from text_label.c.
Dave Wysochanski [Mon, 28 Jun 2010 20:31:01 +0000 (20:31 +0000)]
Allow raw_read_mda_header to be called from text_label.c.

We'd like to pass in mda_header to vgname_from_mda().  In order to
do this, we need to call raw_read_mda_header() from text_label.c,
_text_read(), which gets called from the label_read() path, and
peers into the metadata and update vginfo cache.  We should check
the disable bit here, and if set, not peer into the vg metadata,
thus reducing the I/O to disk.

In the process, move vgname_from_mda() to layout.h, since the fn
only gets called from format_text code, and we need the mda_header
definition from the private layout.h.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoMove dev_open/dev_close outside vgname_from_mda().
Dave Wysochanski [Mon, 28 Jun 2010 20:30:46 +0000 (20:30 +0000)]
Move dev_open/dev_close outside vgname_from_mda().

Refactor vgname_from_mda() so caller must open/close the device.
Should be no functional change.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoMove dev_open / dev_close outside _vg_read_raw_area().
Dave Wysochanski [Mon, 28 Jun 2010 20:30:30 +0000 (20:30 +0000)]
Move dev_open / dev_close outside _vg_read_raw_area().

This refactoring moves the device open/close up one level to the caller of
_vg_read_raw_area().  Should be no functional change and facilitate future
changes related to metadata balancing.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoAdd location independent flag and functions to ignore mdas.
Dave Wysochanski [Mon, 28 Jun 2010 20:30:14 +0000 (20:30 +0000)]
Add location independent flag and functions to ignore mdas.

First we add a 'flags' field to the location independent
metadata_area structure, and a MDA_IGNORE flag.  The
mda_is_ignored and mda_set_ignored functions are added to
manage the flag.  Adding the flag and functions gives a
library interface to ignore metadata areas independent of
the underlying location (disk, file, etc).  The location
specific read/write functions must then handle the specifics
of what this flag means to the location.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
Reviewed-by: Alasdair G Kergon <agk@redhat.com>
14 years agoAdd text format specific 'rlocn' ignore flag and access functions.
Dave Wysochanski [Mon, 28 Jun 2010 20:29:57 +0000 (20:29 +0000)]
Add text format specific 'rlocn' ignore flag and access functions.

Adding a flag to the 'rlocn' structure in the mda header of the
text format allows us to flip a bit to ignore an area on disk that
stores the metadata via the text format specific mda_header.
This patch defines the flag and access functions to manage the flag.
Other patches will manage the ignore on a format-independent basis,
by using a flag in the metadata_area structure.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoChange 'filler' to 'flags' in on-disk 'raw_locn' structure.
Dave Wysochanski [Mon, 28 Jun 2010 20:29:42 +0000 (20:29 +0000)]
Change 'filler' to 'flags' in on-disk 'raw_locn' structure.

Future patches will make use of a specific flag in the on-disk 'raw_locn'
structure to enable/disable metadata areas, and facilitate metadata
balancing.

Note that 'filler' is always set to '0' (see add_mda() - memset),
so use of this area as a non-zero flags field is a safe way to
provide future code features.

Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
14 years agoMinor shell style cleanup.
Petr Rockai [Mon, 28 Jun 2010 19:13:33 +0000 (19:13 +0000)]
Minor shell style cleanup.

14 years agoRefactor the handles_missing_pv logic in lvchange.
Petr Rockai [Mon, 28 Jun 2010 19:10:16 +0000 (19:10 +0000)]
Refactor the handles_missing_pv logic in lvchange.

14 years agoFix for bz608048 from Taka...
Jonathan Earl Brassow [Mon, 28 Jun 2010 14:19:41 +0000 (14:19 +0000)]
Fix for bz608048 from Taka...

The same region size is used for both mirror volume and mirrored
log volume, but when the physical extent size is bigger than region size,
the size of mirror leg for mirrored log is smaller than the region size
and lvcreate command fails.

This patch adjusts a region size of mirrored log to a smaller value of
region size or physical extent size.

[This patch ensures that the region_size of the mirrored log does not
exceed the size of the mirrored log itself, which would violate the
kernel constraint: (region_size <= ti->len).]

Signed-off-by: Takahiro Yasui <takahiro.yasui@hds.com>
Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
14 years agogenerate liblvm2cmd exported symbols too
Alasdair Kergon [Fri, 25 Jun 2010 18:23:10 +0000 (18:23 +0000)]
generate liblvm2cmd exported symbols too

14 years agoGenerate liblvm2app and libdevmapper exported symbols from header files.
Alasdair Kergon [Fri, 25 Jun 2010 18:17:38 +0000 (18:17 +0000)]
Generate liblvm2app and libdevmapper exported symbols from header files.

Detection is simply by prefix - dm_ or lvm_ - and any additional symbols needed
but not detected this way are placed in .exported_symbols.

14 years agoactually, let's keep these in same order as in header
Alasdair Kergon [Fri, 25 Jun 2010 12:21:47 +0000 (12:21 +0000)]
actually, let's keep these in same order as in header

14 years agoUpdate liblvm2app exported symbols.
Alasdair Kergon [Fri, 25 Jun 2010 12:19:52 +0000 (12:19 +0000)]
Update liblvm2app exported symbols.
Add Makefile target to generate current list of lvm2app.h functions.

14 years agoFix typo reported in Debian bugzilla #586043
Zdenek Kabelac [Thu, 24 Jun 2010 08:36:57 +0000 (08:36 +0000)]
Fix typo reported in Debian bugzilla #586043

14 years agoPreload libc locale messages.
Zdenek Kabelac [Thu, 24 Jun 2010 08:29:30 +0000 (08:29 +0000)]
Preload libc locale messages.

Preload libc.mo file for localized lvm before taking memory lock - this way
we prevent disk access for some error paths in libdm, that prints localized
errno messages while they are still in memory locked state.

14 years agoAdd few missing information about what is this script doing.
Zdenek Kabelac [Thu, 24 Jun 2010 08:18:54 +0000 (08:18 +0000)]
Add few missing information about what is this script doing.
(based upon Debian bugzilla suggestion)

14 years agoAdd a test of wait_for_locks behaviour (adapted from an original by Dave).
Petr Rockai [Thu, 24 Jun 2010 07:57:54 +0000 (07:57 +0000)]
Add a test of wait_for_locks behaviour (adapted from an original by Dave).

14 years agoupdate WHATS_NEW file with entry for simultaneous mirror image and
Jonathan Earl Brassow [Wed, 23 Jun 2010 21:01:42 +0000 (21:01 +0000)]
update WHATS_NEW file with entry for simultaneous mirror image and
mirrored log image fault-handling fix.

14 years agoCommitting Taka's patch... He found a problem during
Jonathan Earl Brassow [Wed, 23 Jun 2010 20:32:29 +0000 (20:32 +0000)]
Committing Taka's patch...  He found a problem during
the failure of a device that contained both a image of
a mirror and an image of the mirrored log.  The order
of the handling of those faults was important (and
wrong), this patch corrects that.

Patch-From: Takahiro Yasui <tyasui@redhat.com>

14 years agopost-release
Alasdair Kergon [Wed, 23 Jun 2010 19:35:11 +0000 (19:35 +0000)]
post-release

14 years agopre-release
Alasdair Kergon [Wed, 23 Jun 2010 17:48:41 +0000 (17:48 +0000)]
pre-release

14 years agoIn some C++ standards, typeof is not reserved.
Alasdair Kergon [Wed, 23 Jun 2010 17:03:14 +0000 (17:03 +0000)]
In some C++ standards, typeof is not reserved.

14 years agoFix udev rules to handle spurious events properly.
Peter Rajnoha [Wed, 23 Jun 2010 17:00:32 +0000 (17:00 +0000)]
Fix udev rules to handle spurious events properly.

We can use DM_UDEV_PRIMARY_SOURCE_FLAG to identify the spurious events
and use it as an indication that the device has already been activated before
(and hence we can find this property in udev database).

WARNING: This change requires udev startup script to preserve udev database
from initrd. All the information stored there during activation of devices
is important for the initial "udevadm trigger --action=add" call that is
used in udev startup script. If not done this way, udev startup script needs
to define DM_UDEV_PRIMARY_SOURCE_FLAG=1 property for any ADD events it uses.

14 years agoFix clvmd init script status
Milan Broz [Wed, 23 Jun 2010 16:24:13 +0000 (16:24 +0000)]
Fix clvmd init script status

- s/Active clustred VG/clustered VG/ (only LV can be active)

- print only active LVs (not all) in status command
(In the lvdisplay form /dev/vg/lv.)

For now, still use awk (already used in clustered_vgs).

https://bugzilla.redhat.com/show_bug.cgi?id=598495

14 years agoUse more standard naming for PVs and VG in vgimportclone example.
Mike Snitzer [Wed, 23 Jun 2010 16:12:30 +0000 (16:12 +0000)]
Use more standard naming for PVs and VG in vgimportclone example.

14 years agoCleanup sentences of the example provided in the vgimportclone man page
Mike Snitzer [Wed, 23 Jun 2010 14:15:55 +0000 (14:15 +0000)]
Cleanup sentences of the example provided in the vgimportclone man page
(motivated by a patch that Debian was carrying).

14 years agoThe function that runs to compress a stacked mirror after
Jonathan Earl Brassow [Wed, 23 Jun 2010 13:57:26 +0000 (13:57 +0000)]
The function that runs to compress a stacked mirror after
converting from 2-way to 3-way mirror (collapse_mirrored_lv)
was calling '_remove_mirror_images' with the 'remove_log'
parameter set.  When the code was put in to fix 599898 to
honor log parameters during conversion, this argument was
suddenly being honored.  Thus, when someone would convert from
a 2-way to 3-way mirror, the log would get removed.

'collapse_mirrored_lv' should not be calling '_remove_mirror_images'
with 'remove_log' set.

14 years agoFix typo: "INTERNAL ERROR" -> "INTERNAL_ERROR"
Zdenek Kabelac [Wed, 23 Jun 2010 12:54:46 +0000 (12:54 +0000)]
Fix typo: "INTERNAL ERROR" -> "INTERNAL_ERROR"

Author: Xinwei Hu
        xwhu at novell dot com

14 years agoAdd lv_path to reports to offer full /dev pathname.
Alasdair Kergon [Wed, 23 Jun 2010 12:32:08 +0000 (12:32 +0000)]
Add lv_path to reports to offer full /dev pathname.

14 years agoFix typo: premitted -> permitted
Zdenek Kabelac [Wed, 23 Jun 2010 10:22:59 +0000 (10:22 +0000)]
Fix typo: premitted ->  permitted

Signed-off-by: Takahiro Yasui<tyasui@redhat.com>
14 years agoBetter WHATS_NEW message - as we still have not fixed 'make exec_prefix' build
Zdenek Kabelac [Wed, 23 Jun 2010 10:21:00 +0000 (10:21 +0000)]
Better WHATS_NEW message - as we still have not fixed 'make exec_prefix' build

14 years agoFix "allocated" warning typo.
Milan Broz [Tue, 22 Jun 2010 21:10:53 +0000 (21:10 +0000)]
Fix "allocated" warning typo.

14 years agoAdd device name to output of error messages in raw_read_mda_header().
Dave Wysochanski [Tue, 22 Jun 2010 19:18:27 +0000 (19:18 +0000)]
Add device name to output of error messages in raw_read_mda_header().

It would be helpful if we had the device name when something like
a mda_header checksum error occurs.

Before:
./tools/lvm pvs -opv_name,vg_name,uuid,mda_count,pv_mda_count_ignored,vg_mda_count,vg_mda_count_ignored,vg_mda_copies
  Incorrect metadata area header checksum
  PV         VG      PV UUID                                #PMda #PMdaIgn #VMda #VMdaIgn #VMdaCps
  /dev/loop0 vgtest2 sVv26t-gjpb-Rcau-uBDO-Cx04-GbRR-6Ssq7e     2        0     4        0        4
  /dev/loop1 vgtest2 zXWStT-qE8F-mbkc-RfgH-aytv-mptF-Y5Ce09     2        0     4        0        4
  /dev/loop2         riCpK9-9G8r-LlIp-i2oh-mb3N-CUzk-u5YpuR     1        0     0        0        0
  /dev/loop3 vgtest  tQCUjm-rmyd-i92d-4eeE-UYBW-v1vQ-kRaA17     2        0     4        2        0
  /dev/loop4 vgtest  ZRvpeI-p8F1-ccVW-BBac-xhl1-aGXU-CbP0oo     2        2     4        2        0

After:
./tools/lvm pvs -opv_name,vg_name,uuid,mda_count,pv_mda_count_ignored,vg_mda_count,vg_mda_count_ignored,vg_mda_copies
  Incorrect metadata area header checksum on /dev/loop2 at offset 4096
  PV         VG      PV UUID                                #PMda #PMdaIgn #VMda #VMdaIgn #VMdaCps
  /dev/loop0 vgtest2 sVv26t-gjpb-Rcau-uBDO-Cx04-GbRR-6Ssq7e     2        0     4        0        4
  /dev/loop1 vgtest2 zXWStT-qE8F-mbkc-RfgH-aytv-mptF-Y5Ce09     2        0     4        0        4
  /dev/loop2         riCpK9-9G8r-LlIp-i2oh-mb3N-CUzk-u5YpuR     1        0     0        0        0
  /dev/loop3 vgtest  tQCUjm-rmyd-i92d-4eeE-UYBW-v1vQ-kRaA17     2        0     4        2        0
  /dev/loop4 vgtest  ZRvpeI-p8F1-ccVW-BBac-xhl1-aGXU-CbP0oo     2        2     4        2        0

14 years agoUse flexible data[] in cmirrord request to prevent abort in runtime size checks.
Milan Broz [Tue, 22 Jun 2010 13:11:29 +0000 (13:11 +0000)]
Use flexible data[] in cmirrord request to prevent abort in runtime size checks.

14 years agoAdding section number for lvm.conf manpages.
Zdenek Kabelac [Tue, 22 Jun 2010 07:34:34 +0000 (07:34 +0000)]
Adding section number for lvm.conf manpages.

14 years agoMirrors can be layered - as in the case of an converting 2-way
Jonathan Earl Brassow [Mon, 21 Jun 2010 16:12:33 +0000 (16:12 +0000)]
Mirrors can be layered - as in the case of an converting 2-way
to 3-way mirror.  When conversion operations are performed on
these types of mirrors, log options can be confused/ignored.

In the case of a converting 3-way mirror, we have a top-level
2-way corelog mirror whose legs are 1) a 2-way disk-log mirror
and 2) a linear device.  If we wish to convert this 3-way mirror
to a 2-way mirror, the linear device is removed and the extra
top layer is eliminated.  If we also wished to convert the disk
log to a core log in the same step, ambiguity creeps in.  It is
somewhat obvious what the user wants - a 2-way mirror with a
corelog.  However, looking at the top level mirror before
compression, it seems that the mirror already has a core log.
This is why the operation seemed to fail.

This patch simply re-evaluates what mirrored_seg points to after
a compression and then considers the log argument.

This is a fix for bug 599898.

14 years agoAdd error checking for calls to sprintf - it can fail for more
Jonathan Earl Brassow [Mon, 21 Jun 2010 16:07:06 +0000 (16:07 +0000)]
Add error checking for calls to sprintf - it can fail for more
reasons than just 'out-of-space'.

14 years agoVarious cleanups following recent commits.
Alasdair Kergon [Mon, 21 Jun 2010 15:56:57 +0000 (15:56 +0000)]
Various cleanups following recent commits.

14 years agoLet running clvmd process time to reexec.
Milan Broz [Mon, 21 Jun 2010 10:45:15 +0000 (10:45 +0000)]
Let running clvmd process time to reexec.

Because execve stops the command loop,
we never receive response (only socket close) for clvmd -S,
so waiting for response here makes no sense.

But if the calling process (clvmd -S)  exits too early, connection
is closed from client side, clvmd takes this as an error and
never run restart code.

Ugly hack(TM).

14 years agoUse early udev synchronisation and update of dev nodes for clustered mirrors.
Peter Rajnoha [Mon, 21 Jun 2010 08:54:32 +0000 (08:54 +0000)]
Use early udev synchronisation and update of dev nodes for clustered mirrors.

When using clustered mirrors, we need device nodes to be created during
processing of device tree, not at its end like we normally do (we need to
access the nodes in cmirror prematurely). Therefore we use a new flag called
"immediate_dev_node" stored in deptree's load_properties struct to instruct the
device tree processing code to immediately synchronize with udev and flush all
stacked node operations so the nodes are prepared for use.

For now, the immediate_dev_node is used for clustered mirrors during
processing the dm_tree_preload_children code only. We can add more later if
needed.

14 years agodaemons/cmirrord/functions.c (part of cmirrord) was referencing
Jonathan Earl Brassow [Fri, 18 Jun 2010 20:58:04 +0000 (20:58 +0000)]
daemons/cmirrord/functions.c (part of cmirrord) was referencing
linux/kdev_t.h even though it wasn't needed.  Strangely, it seems
to be causing problems on various architectures (i686) in the
function daemons/cmirrord/functions.c:disk_status_info()->sprintf.

I'm not sure why this is a problem since none of the macros in
kdev_t.h are used in that code, but it certainly doesn't hurt to
pull an unnecessary header and it seems to fix the problem.

14 years agoAdd man pages for lvmconf and unsupported lvmsadc and lvmsar tools.
Zdenek Kabelac [Fri, 18 Jun 2010 10:19:29 +0000 (10:19 +0000)]
Add man pages for lvmconf and unsupported lvmsadc and lvmsar tools.

14 years agoFix exit code when requesting help using documented -o help option.
Milan Broz [Thu, 17 Jun 2010 13:15:51 +0000 (13:15 +0000)]
Fix exit code when requesting help using documented -o help option.

IOW fix lvs -o help
...
  Command failed with status code 5.

14 years agoClean up cluster lock mode and flags definition.
Milan Broz [Thu, 17 Jun 2010 12:48:54 +0000 (12:48 +0000)]
Clean up cluster lock mode and flags definition.

Code is mixing up internal DLM and LVM definitions of lock
modes and flags.

OpenAIS and singlenode locking do not depend on DLM but
code currently cannot be compiled without libdlm.h!

LCK_* flags is LVM abstraction, used through all the code.
Only low-level backend (clvmd-cman etc) should use DLM definitions,
also this code should do all needed conversions.

Because there are two DLM flags used in generic code
(NOQUEUE, CONVERT) we define it similar way like lock modes.
(So all needed binary-compatible flags are on one place in locking.h)

(Further code cleaning still needed, though:-)

14 years agoAdd man page for dmeventd
Zdenek Kabelac [Thu, 17 Jun 2010 12:14:43 +0000 (12:14 +0000)]
Add man page for dmeventd
(inspired by patch from Danny Rawlins, monster.romster at gmail dot com)

14 years agoUpdate lvresize/extend/reduce manpages with --nofsck, --resizefs options
Zdenek Kabelac [Thu, 17 Jun 2010 11:25:43 +0000 (11:25 +0000)]
Update lvresize/extend/reduce manpages with --nofsck, --resizefs options

14 years agoFix lvm2cmd example in documentation.
Milan Broz [Wed, 16 Jun 2010 13:03:48 +0000 (13:03 +0000)]
Fix lvm2cmd example in documentation.

14 years agoRemove C++ private keyword from headers.
Milan Broz [Wed, 16 Jun 2010 13:01:25 +0000 (13:01 +0000)]
Remove C++ private keyword from headers.
Add extern C definition for libdevmapper, lvm2app and lvm2cmd.

14 years agoUse "" instead of <> for configure.h and libdevmapper.h
Zdenek Kabelac [Tue, 15 Jun 2010 11:00:44 +0000 (11:00 +0000)]
Use "" instead of <> for  configure.h and libdevmapper.h
Move configure.h as the first header for clvmd source files.

14 years agoUpdate WHATS_NEW for last commit
Zdenek Kabelac [Mon, 7 Jun 2010 14:39:18 +0000 (14:39 +0000)]
Update WHATS_NEW  for last commit

14 years agoFix wrong usage of exec_prefix from previous patch introducing LVM_PATH define
Zdenek Kabelac [Mon, 7 Jun 2010 14:31:59 +0000 (14:31 +0000)]
Fix wrong usage of exec_prefix from previous patch introducing LVM_PATH define

Introduce lvm_exec_prefix with resolved exec_prefix.
(using same ac_default_prefix as for CLVMD_PATH)
Use lvm_exec_prefix instead of dmeventd_prefix (fixes missing ac_default_prefix)

Note: This patch is rather hot-fix as currently generate code
does not create correct code for make exec_prefix=

14 years agoFix segfault in clvmd -R if no response from daemon received.
Milan Broz [Mon, 7 Jun 2010 13:52:52 +0000 (13:52 +0000)]
Fix segfault in clvmd -R if no response from daemon received.

14 years agopost-release
Alasdair Kergon [Mon, 7 Jun 2010 10:25:43 +0000 (10:25 +0000)]
post-release

14 years agopre-release
Alasdair Kergon [Fri, 4 Jun 2010 17:23:49 +0000 (17:23 +0000)]
pre-release

14 years agoFix restart of clvmd using -S switch
Milan Broz [Fri, 4 Jun 2010 12:59:30 +0000 (12:59 +0000)]
Fix restart of clvmd using -S switch

 - allocate environment dynamically (still missing some limit?)
 - try to recover, if destroy failed (do not destroy lvm here) and free memory
 - check strdup() return codes
 - report failure to log
 - do not print NULL in exclusive lock loop

14 years agoFix clvmd initscript restart command to start clvmd if not yet running.
Milan Broz [Thu, 3 Jun 2010 21:03:53 +0000 (21:03 +0000)]
Fix clvmd initscript restart command to start clvmd if not yet running.

14 years agoUse INSTALL_DIR to create directories
Zdenek Kabelac [Thu, 3 Jun 2010 13:52:21 +0000 (13:52 +0000)]
Use INSTALL_DIR to create directories

14 years agoUse absolute paths in commands
Zdenek Kabelac [Thu, 3 Jun 2010 13:50:26 +0000 (13:50 +0000)]
Use absolute paths in commands

clvmd restart does not work at all if clvmd binary is not in current
dir.

14 years agoRequire partial option in lvchange --refresh for partials LVs.
Milan Broz [Thu, 3 Jun 2010 12:45:05 +0000 (12:45 +0000)]
Require partial option in lvchange --refresh for partials LVs.

We must not refresh LV if some PVs are missing and partial activation
was not requested.

Fixes https://bugzilla.redhat.com/show_bug.cgi?id=598886

14 years agoRevert _init_rand() to reset errno - restores original init behavior.
Dave Wysochanski [Tue, 1 Jun 2010 21:47:57 +0000 (21:47 +0000)]
Revert _init_rand() to reset errno - restores original init behavior.

14 years agoDo not fail lvm_init() if init_logging() generates an errno.
Dave Wysochanski [Tue, 1 Jun 2010 21:46:29 +0000 (21:46 +0000)]
Do not fail lvm_init() if init_logging() generates an errno.

Revert to original behavior of lvm commands if init_logging() generated an
errno.  Fixes rhbz 592967.

14 years agoDon't merge unchanged persistent cache file before dumping if tool scanned.
Alasdair Kergon [Tue, 1 Jun 2010 19:02:12 +0000 (19:02 +0000)]
Don't merge unchanged persistent cache file before dumping if tool scanned.

14 years agoAdd support for dm-mod module autoload.
Peter Rajnoha [Tue, 1 Jun 2010 16:08:13 +0000 (16:08 +0000)]
Add support for dm-mod module autoload.

A kernel patch is on its way for 2.6.35 adding support for dm-mod module
autoload. Udev v155 and higher is able to read static node information given
in modules.devname (extracted by depmod before) and will create such nodes
at its start. The first access to such node will load the module automatically
(directly in kernel) before the actual read/write operation is processed.

14 years agoFix incorrect memory pool deallocation while using vg_read for files.
Peter Rajnoha [Tue, 1 Jun 2010 12:08:50 +0000 (12:08 +0000)]
Fix incorrect memory pool deallocation while using vg_read for files.

We create a separate pool "lvm2 vg_read" for vg_read and we don't use
cmd->mem anymore.

14 years agoAdd --type parameter description to the lvcreate man page.
Mike Snitzer [Fri, 28 May 2010 03:50:15 +0000 (03:50 +0000)]
Add --type parameter description to the lvcreate man page.

14 years agoDocument 'clear' in dmsetup man page.
Mike Snitzer [Thu, 27 May 2010 19:00:14 +0000 (19:00 +0000)]
Document 'clear' in dmsetup man page.

14 years agoUse expected union semun for arguments in selected semaphore operations.
Peter Rajnoha [Thu, 27 May 2010 15:02:56 +0000 (15:02 +0000)]
Use expected union semun for arguments in selected semaphore operations.

This is standard and expected use. It also prevents errors related with
misalignment of the arguments for semaphore operations on some architectures.

14 years agoFix copy&paste detection of kernel release version.
Zdenek Kabelac [Tue, 25 May 2010 08:40:36 +0000 (08:40 +0000)]
Fix copy&paste detection of kernel release version.
Add log_error to avoid return_0 without log_error.

14 years agoReplace strncmp kernel version number checks with proper ones
Alasdair Kergon [Mon, 24 May 2010 23:11:34 +0000 (23:11 +0000)]
Replace strncmp kernel version number checks with proper ones

14 years agoAvoid selecting names under /dev/block if there is an alternative.
Alasdair Kergon [Mon, 24 May 2010 22:53:48 +0000 (22:53 +0000)]
Avoid selecting names under /dev/block if there is an alternative.

14 years agoFix topology test after last changes to not leak loop devices.
Milan Broz [Mon, 24 May 2010 19:27:38 +0000 (19:27 +0000)]
Fix topology test after last changes to not leak loop devices.

14 years agoChoose between clustered log versions based on kernel version.
Alasdair Kergon [Mon, 24 May 2010 17:46:47 +0000 (17:46 +0000)]
Choose between clustered log versions based on kernel version.
Add fixmes for broken strcmp.

14 years agoUse mv -f for replace config in test.
Milan Broz [Mon, 24 May 2010 17:19:50 +0000 (17:19 +0000)]
Use mv -f for replace config in test.

14 years agoReplace lvm.conf instead of truncating it.
Milan Broz [Mon, 24 May 2010 17:18:05 +0000 (17:18 +0000)]
Replace lvm.conf instead of truncating it.

14 years agot-lvconvert-repair-transient hangs on .33 and on RHEL6 kernel -> disable
Petr Rockai [Mon, 24 May 2010 16:33:21 +0000 (16:33 +0000)]
t-lvconvert-repair-transient hangs on .33 and on RHEL6 kernel -> disable

14 years agoUpdate clustered log kernel module name to log-userspace.
Alasdair Kergon [Mon, 24 May 2010 16:30:15 +0000 (16:30 +0000)]
Update clustered log kernel module name to log-userspace.

14 years agoAccount for mirror transient status when doing lvconvert --repair.
Petr Rockai [Mon, 24 May 2010 15:32:20 +0000 (15:32 +0000)]
Account for mirror transient status when doing lvconvert --repair.

14 years agoSkip t-pvcreate-operation-md if mdadm fails to create the required device.
Petr Rockai [Mon, 24 May 2010 15:28:45 +0000 (15:28 +0000)]
Skip t-pvcreate-operation-md if mdadm fails to create the required device.

14 years agoUpdate Copyright date for resently modifed files
Zdenek Kabelac [Mon, 24 May 2010 09:04:27 +0000 (09:04 +0000)]
Update Copyright date for resently modifed files

14 years agoReplicator: update activate code for vgchange
Zdenek Kabelac [Mon, 24 May 2010 09:03:39 +0000 (09:03 +0000)]
Replicator: update activate code for vgchange

Activate only the first replicator-dev LV, that activates all other
related LVs from Replicator. In case of error during this activation,
it will not retry again for other heads (less confusing error log).

14 years agoReplicator: add replicator to dtree
Zdenek Kabelac [Mon, 24 May 2010 09:01:05 +0000 (09:01 +0000)]
Replicator: add replicator to dtree

Adding all replicator related LVs to dtree.
Start of one replicator_dev initiate start of all other related.

14 years agoReplicator: VG with cmd_missing_vgs does not generate output
Zdenek Kabelac [Mon, 24 May 2010 08:59:29 +0000 (08:59 +0000)]
Replicator: VG with cmd_missing_vgs does not generate output

Do not print message if missing VG is found.

14 years agoFix scripts/relpath.awk to work with mawk
Mikulas Patocka [Fri, 21 May 2010 15:28:16 +0000 (15:28 +0000)]
Fix scripts/relpath.awk to work with mawk

length(array) is specific to GNU awk and doesn't work in mawk.
Use a return value of "split" function to indicate array size, this is
supported in both gawk and mawk.

This patch fixes the following errors during "make install" when mawk is
installed as a default awk.

mawk: scripts/relpath.awk: line 25: illegal reference to array from
mawk: scripts/relpath.awk: line 25: illegal reference to array to
mawk: scripts/relpath.awk: line 27: illegal reference to array from
mawk: scripts/relpath.awk: line 32: illegal reference to array to

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
14 years agoReplicator: update _create_partial_dtree()
Zdenek Kabelac [Fri, 21 May 2010 14:47:58 +0000 (14:47 +0000)]
Replicator: update _create_partial_dtree()

Adding function _add_partial_replicator_to_dtree() to create
partial tree for Replicator target.

Using dm_tree_node_set_presuspend_node() for Replicator.

14 years agoReplicator: activate checks for missing vgs
Zdenek Kabelac [Fri, 21 May 2010 14:34:01 +0000 (14:34 +0000)]
Replicator: activate checks for missing vgs

Activation needs to have all remote VGs locked to pass for replicator.
So if activated LV is replicator-dev LV - read all remote VGs.

14 years agoReplicator: lock_lv_vol() finds missing VGs
Zdenek Kabelac [Fri, 21 May 2010 14:29:49 +0000 (14:29 +0000)]
Replicator: lock_lv_vol() finds missing VGs

Find and check for all needed VGs before calling lock_vol().

14 years agoReplicator: use cmd_vg for process_each_lv_in_vg()
Zdenek Kabelac [Fri, 21 May 2010 14:15:39 +0000 (14:15 +0000)]
Replicator: use cmd_vg for process_each_lv_in_vg()

As for _process_one_vg() we need similar retry loop for
process_each_lv_in_vg(). This patch retries to process
failed LVs with reopened VGs.

Patch does not add any extra repeated invocations if there is not
found any missing VG during LV processing.

14 years agoReplicator: use cmd_vg list for _process_one_vg()
Zdenek Kabelac [Fri, 21 May 2010 14:11:13 +0000 (14:11 +0000)]
Replicator: use cmd_vg list for _process_one_vg()

Patch modifes behavior of _process_one_vg().

In the first pass vg_read() collectis for replicator sorted list of
additional VGs during lock_vol().
If any other VG is needed by the replicator and it is not yet opened
then next iteration loop is taken with all collected VGs.

Flag vg->cmd_missing_vgs detects missing VGs.

14 years agoReplicator: add read and release VGs for rsites
Zdenek Kabelac [Fri, 21 May 2010 14:07:16 +0000 (14:07 +0000)]
Replicator: add read and release VGs for rsites

Add functions to read and release remote VGs for replicator sites
in activation context.

14 years agoAdd toolcontext.h header file.
Zdenek Kabelac [Fri, 21 May 2010 13:34:09 +0000 (13:34 +0000)]
Add toolcontext.h header file.

14 years agoRemove files from wrong directory
Zdenek Kabelac [Fri, 21 May 2010 13:18:25 +0000 (13:18 +0000)]
Remove files from wrong directory

14 years agoRight directory name for replicator files.
Zdenek Kabelac [Fri, 21 May 2010 13:17:20 +0000 (13:17 +0000)]
Right directory name for replicator files.

My local CVS was placing file in slightly different directory by using
obsolete files.

14 years agoHmm - fixing cvs import mistake
Zdenek Kabelac [Fri, 21 May 2010 13:14:02 +0000 (13:14 +0000)]
Hmm - fixing cvs import mistake

14 years agoReplicator: add find_replicator_vgs
Zdenek Kabelac [Fri, 21 May 2010 12:55:25 +0000 (12:55 +0000)]
Replicator: add find_replicator_vgs

Adding find_replicator_vgs() function to find all needed
VGs for replicator-dev LV.

This function is later called before taking lock_vol().

14 years agoReplicator: add sorted cmd_vg list
Zdenek Kabelac [Fri, 21 May 2010 12:52:01 +0000 (12:52 +0000)]
Replicator: add sorted cmd_vg list

Introduce struct cmd_vg to store information about needed
volume group name, vgid, flags and the pointer to opened VG.

Keep VGs list in alphabetical order for locking order.

Introduce functions:
cmd_vg_add() add new cmd_vg entry.
cmd_vg_lookup() search cmd_vgs for vg_name.
cmd_vg_read() open VGs in cmd_vgs list.
cmd_vg_release() close VGs in reversed order.

14 years agoReplicator: extend volume_group with list of VGs and flag
Zdenek Kabelac [Fri, 21 May 2010 12:47:46 +0000 (12:47 +0000)]
Replicator: extend volume_group with list of VGs and flag

Add pointer to linked list of opened VGs. List temporarily keeps
the information about needed or locked and opened VGs for replicator target.

Also add cmd_missing_vgs flag information for quick check and
also for possible continuos process_each_lv() usage where we need
to detect whether failure has been caused by missing VG or
some other reason.

This page took 0.071215 seconds and 5 git commands to generate.