]>
git.ipfire.org Git - thirdparty/mdadm.git/log
Neil Brown [Thu, 14 Dec 2006 06:31:29 +0000 (17:31 +1100)]
Fix and test --update=uuid
A number of odd bugs here, but now we have a regression test as well.
Neil Brown [Thu, 14 Dec 2006 06:31:25 +0000 (17:31 +1100)]
When resync finished, report the mismatch count if there is one.
This doesn't get mailed out, but will appear in syslog...
Maybe it should be mailed if it was a 'check' or 'repair' pass...
Neil Brown [Thu, 14 Dec 2006 06:31:22 +0000 (17:31 +1100)]
--wait or -W will wait for resync activity to finish on the given devices.
Neil Brown [Thu, 14 Dec 2006 06:31:19 +0000 (17:31 +1100)]
Change 'Device Size' to 'Used Dev Size'
because it only shows how much of each device is actually used, not
how big they are.
Neil Brown [Thu, 14 Dec 2006 06:31:16 +0000 (17:31 +1100)]
Increase raid456 stripe cache size if needed to --grow the array.
The setting used unfortunately requires intimate knowledge of the
kernel, and it not reset when the reshape finishes.
Neil Brown [Thu, 14 Dec 2006 06:31:13 +0000 (17:31 +1100)]
Give useful message if raid4/5/6 cannot be started because it is not clean and is also degraded.
Neil Brown [Thu, 14 Dec 2006 06:31:10 +0000 (17:31 +1100)]
Default to --auto=yes
so the array devices with 'standard' names
get created automatically, as this is almost always what is wanted.
Neil Brown [Thu, 14 Dec 2006 06:31:08 +0000 (17:31 +1100)]
Fix a misleading comment.
Neil Brown [Thu, 14 Dec 2006 06:31:03 +0000 (17:31 +1100)]
Support --update=devicesize for cases where the underlying device can change size.
Neil Brown [Thu, 14 Dec 2006 06:31:00 +0000 (17:31 +1100)]
Improve allocation and use of space for bitmaps in version1 metadata
Depending on the size of the array we reserve space for up to 128K
of bitmap, and we use it where possible.
When hot-adding to a version 1.0 we can still only use the 3K at the
end though - need a sysfs interface to improve that.
If a small chunksize is requested on Create, we don't auto-enlarge
the reserved space - this still needs to be fixed.
Neil Brown [Thu, 14 Dec 2006 06:30:56 +0000 (17:30 +1100)]
Actually support --syslog
Fixes-debian-bug: 402457 (part)
All the code is there, and it works for '-y', but for some reason,
"syslog" was missing from the long_options.
Neil Brown [Thu, 14 Dec 2006 06:30:53 +0000 (17:30 +1100)]
Remove some unused interfaces to the metadata handlers.
This stuff has never been used (at least as far as git history can
see). I wonder why it was there...
Neil Brown [Thu, 14 Dec 2006 06:30:51 +0000 (17:30 +1100)]
Make Assemble/Force work on raid6 with 2 missing devices.
Previously it onl worked when one missing device.
Also split the "force" update_super method into two and it
is really serving two functions.
Neil Brown [Sun, 12 Nov 2006 22:12:09 +0000 (09:12 +1100)]
--update=resync did exactly the wrong thing for version1 metadata.
Neil Brown [Thu, 9 Nov 2006 05:28:27 +0000 (16:28 +1100)]
Fixed UUID printing in "--detail --brief" for version1 metadata.
Neil Brown [Wed, 8 Nov 2006 23:00:32 +0000 (10:00 +1100)]
Release 2.5.6
Neil Brown [Wed, 8 Nov 2006 22:38:50 +0000 (09:38 +1100)]
Fixed problems that could cause infinitel loop with auto assemble.
If an auto-assembly attempt failes because the array cannot be
opened or because the array has already been created, then we
get into an infinite loop.
Reported-by: Dan Pascu <dan@ag-projects.com>
Fixes-debian-bug: 396582
Goswin Brederlow [Wed, 8 Nov 2006 22:37:45 +0000 (09:37 +1100)]
Document that the default metadata version can be overridden in config file.
From: Goswin Brederlow <brederlo@informatik.uni-tuebingen.de>
This is already mentioned in the config documentation, but not in the
place when the normal default is mentioned.
Fixes-debian-bug: 396914
Neil Brown [Wed, 8 Nov 2006 22:37:10 +0000 (09:37 +1100)]
Strdup the bitmap file name found in the config file
Otherwise we quickly lose it...
Neil Brown [Fri, 3 Nov 2006 03:52:28 +0000 (14:52 +1100)]
Release 2-5-5
Neil Brown [Mon, 23 Oct 2006 05:51:26 +0000 (15:51 +1000)]
Support --examine --brief --verbose properly
Similar to -Esv, this combination should print out the
'devices=' line in the otherwise --breif output.
Michael Labuschke [Mon, 23 Oct 2006 04:56:35 +0000 (14:56 +1000)]
Check error return on a couple of syscalls.
From: "Michael Labuschke" <michael@labuschke.de>
Neil Brown [Mon, 23 Oct 2006 04:56:22 +0000 (14:56 +1000)]
Update test scripts for new blockdev
--getsize64 seems to have disappeared.
Neil Brown [Sun, 22 Oct 2006 22:56:27 +0000 (08:56 +1000)]
Improve error message when wrong --update option is given.
Neil Brown [Thu, 19 Oct 2006 06:46:38 +0000 (16:46 +1000)]
Increase default size of bitmap for v1 superblocks.
It can be increased further, but that takes a bit more code
so will wait for 2.6.
Neil Brown [Thu, 19 Oct 2006 06:38:29 +0000 (16:38 +1000)]
Fix bugs related to raid10 and the new offset layout.
Need to mask of bits above the bottom 16 when calculating number of
copies.
Neil Brown [Thu, 19 Oct 2006 06:38:20 +0000 (16:38 +1000)]
Fix array-subscript error.
->name is sized to '33' to have room for a trailing nul,
but thhat needs to get put in position '32'. Doh!
Luca Berra [Mon, 16 Oct 2006 05:27:48 +0000 (15:27 +1000)]
Don't trigger and error on -As if all arrays are already started.
From: Luca Berra <bluca@comedia.it>
Put another way, if we find any active arrays, then assume things are
going as planned.
Luca Berra [Mon, 16 Oct 2006 05:27:16 +0000 (15:27 +1000)]
Don't fail md_open if array already active.
From: Luca Berra <bluca@comedia.it>
There is no need and it causes occasional problems.
Luca Berra [Mon, 16 Oct 2006 05:26:53 +0000 (15:26 +1000)]
Mdassemble improvements
From: Luca Berra <bluca@comedia.it>
- Fix a bug where mdassemble didn't close a filedescriptor and so couldn't assembele
stacked arrays.
- Allow mdassemble, when run a second time, to mark all arrays as writable.
This is useful if they are started read-only as is best at boot-time.
Neil Brown [Mon, 16 Oct 2006 05:26:43 +0000 (15:26 +1000)]
Reduce max bitmap size when bitmap is in a file.
To ensure kernel never needs to allocate more than 1 page contiguously.
Neil Brown [Mon, 16 Oct 2006 04:39:06 +0000 (14:39 +1000)]
Flush bitmap file before truncate.
Otheriwse the last write happens on fclose which is after ftruncate,
to the truncate has no effect.
Neil Brown [Mon, 16 Oct 2006 04:38:13 +0000 (14:38 +1000)]
Don't #include blkpg.h
Just include the content directly. It is safer.
Neil Brown [Fri, 13 Oct 2006 00:12:47 +0000 (10:12 +1000)]
Release 2-5-4
Neil Brown [Thu, 12 Oct 2006 23:02:35 +0000 (09:02 +1000)]
Remove partitions from components of an md array
They do nothing but cause confusion.
Neil Brown [Thu, 12 Oct 2006 22:23:16 +0000 (08:23 +1000)]
Allow a number after --auto=yes
So if the device name is /dev/md/d0, then the number of
partitions is as given. This is useful in 'CREATE' in
mdadm.conf
Neil Brown [Mon, 9 Oct 2006 21:06:24 +0000 (07:06 +1000)]
Improve the message when mdadm detects similar superblocks
If they are for a partition and a whole device (common case)
they old message doesn't really cover the situation. So add
the "overlap" option to the text.
Also detect whether the device list was in mdadm.conf and
act accordingly.
Neil Brown [Mon, 9 Oct 2006 21:06:08 +0000 (07:06 +1000)]
Clarify when update=super-minor happens automatcially
.. in man page
Neil Brown [Mon, 9 Oct 2006 20:30:07 +0000 (06:30 +1000)]
Fix typo in earlier patch.
Thanks Martin Krafft
Paul Clements [Mon, 9 Oct 2006 01:17:22 +0000 (11:17 +1000)]
Fixed endian problem with bitmap metdata
From: Paul Clements <paul.clements@steeleye.com>
Neil Brown [Mon, 9 Oct 2006 01:17:12 +0000 (11:17 +1000)]
Fix some endian-ness issues with v1 superblocks.
Neil Brown [Mon, 9 Oct 2006 01:16:56 +0000 (11:16 +1000)]
Make --examine report chunk size when it should
Currently it doesn't for raid6 and raid10
Neil Brown [Mon, 9 Oct 2006 01:16:33 +0000 (11:16 +1000)]
Minor man page and comment fixes
Thanks To: "Scott Weikart" <Scott.W@Benetech.org>
Neil Brown [Fri, 11 Aug 2006 08:00:05 +0000 (18:00 +1000)]
Allow symlink creation to be disabled from command line or mdadm.conf
Neil Brown [Fri, 11 Aug 2006 08:00:00 +0000 (18:00 +1000)]
Create symlinks from /dev into /dev/md
When (auto)creating devices in /dev/md/, create
an appropriate symlink from /dev to avoid confusion.
Neil Brown [Fri, 11 Aug 2006 07:59:57 +0000 (17:59 +1000)]
Minor improvement in name auto-chosen for version-1 superblocks.
/dev/md_d0 should have name 'd0' by default to
make /dev/md/d0.
Neil Brown [Fri, 11 Aug 2006 07:59:53 +0000 (17:59 +1000)]
Remove indeterminism from names in --examine --brief output.
It is much better for this sort of thing to be predictable
rather than depending on what devices exist, especially as
these days entries in /dev/ often don't pre-exist.
So make it always
/dev/mdX
for version0 superblocks.
Version1 are always /dev/md/NAME
Neil Brown [Mon, 7 Aug 2006 02:00:30 +0000 (12:00 +1000)]
Release 2.5.3
Neil Brown [Mon, 7 Aug 2006 01:30:02 +0000 (11:30 +1000)]
Tidyup automatic name choice for v-1 arrays
/dev/md_d0 now becomes '0', not '_d0'.
Neil Brown [Mon, 7 Aug 2006 01:13:01 +0000 (11:13 +1000)]
Fix starting of degraded arrays.
Recent change broke assembling of degraded arrays, making
it require --run. This fixes that.
Neil Brown [Mon, 7 Aug 2006 01:12:25 +0000 (11:12 +1000)]
Explicitly state GPL license in man pages.
Neil Brown [Mon, 7 Aug 2006 01:11:42 +0000 (11:11 +1000)]
Document usage of version0.91 superblock.
Neil Brown [Tue, 27 Jun 2006 11:31:58 +0000 (21:31 +1000)]
Release 2.5.2
Description...
Neil Brown [Mon, 26 Jun 2006 05:11:01 +0000 (15:11 +1000)]
More consistent honoring of --configfile
Never use /etc/mdadm.conf if --config file is given (previously
some code used one, some used the other).
Neil Brown [Mon, 26 Jun 2006 05:10:55 +0000 (15:10 +1000)]
Fix user-after-free bug in error path in --monitor mode.
Neil Brown [Mon, 26 Jun 2006 05:10:48 +0000 (15:10 +1000)]
Improve ftw handling.
If not 'ftw' is available, still allow openning of devices by dev number.
More recent version of uclibc support nftw, so add support to check
for that.
Neil Brown [Mon, 26 Jun 2006 02:26:12 +0000 (12:26 +1000)]
Fix a recently introduced bug, and make --assemble more resilient to it.
Make -assemble a bit more resilient to finding strange
information in superblocks.
Don't claim newly added spares are InSync!! (don't know why that
code was ever in there)
Neil Brown [Mon, 26 Jun 2006 02:26:09 +0000 (12:26 +1000)]
Work around bug in --add handling for version-1 superblocks
In 2.6.17 (and prior), the dev_number is ignored when a device
is added to an active array. Rather the first free number is used.
So we work around this by making sure we use the first free
number for dev_number.
Description...
Neil Brown [Mon, 26 Jun 2006 02:26:01 +0000 (12:26 +1000)]
Add 'Array Slot' line to --examine for version-1 superblocks
to make it a bit easier to see what is happening.
martin f krafft [Mon, 26 Jun 2006 02:07:12 +0000 (12:07 +1000)]
Fix manpage typo
From: martin f krafft <madduck@madduck.net>
Neil Brown [Mon, 26 Jun 2006 02:06:01 +0000 (12:06 +1000)]
Get the changelog uptodate for recent changes.
Paul Clements [Tue, 20 Jun 2006 00:01:47 +0000 (10:01 +1000)]
Move a variable declaration to the declaration area.
While declaring variables in the middle of code withs with
newer gcc's it doesn't work with older, and it is arguably
less readable, so just do the right thing.
From: Paul Clements <paul.clements@steeleye.com>
Neil Brown [Tue, 20 Jun 2006 00:01:43 +0000 (10:01 +1000)]
Stop map_dev from returning [0:0]
We sometimes need the NULL when major==minor==0.
So make sure all callers of map_dev can cope with NULL.
Paul Clements [Tue, 20 Jun 2006 00:01:23 +0000 (10:01 +1000)]
Fix problem with post-increment usage in macro
Bad/bad/bad, and cause compiler error on ppc (gcc 3.2.3).
From: Paul Clements <paul.clements@steeleye.com>
Neil Brown [Fri, 16 Jun 2006 00:54:14 +0000 (10:54 +1000)]
Release 2.5.1
Neil Brown [Fri, 16 Jun 2006 00:53:55 +0000 (10:53 +1000)]
Update Changelog for 2.5.1
Neil Brown [Fri, 16 Jun 2006 00:53:50 +0000 (10:53 +1000)]
Fix offsetof macro for 64bit hosts
Neil Brown [Fri, 16 Jun 2006 00:26:12 +0000 (10:26 +1000)]
Fix memory leak in monitor mode
When rescanning /dev, we didn't free the old list.
Also don't search for device with a number of 0,0
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 16 Jun 2006 00:21:11 +0000 (10:21 +1000)]
Fix up some internalbitmap tests
martin f. krafft [Fri, 2 Jun 2006 10:53:02 +0000 (20:53 +1000)]
Allow CONFFILE2 to be overridden during make invocation.
martin f krafft [Fri, 2 Jun 2006 08:31:48 +0000 (10:31 +0200)]
CPPFLAGS gets used automatically on implicit rules.
Take it out for cosmetic reasons to prevent duplicate -DDEBIAN.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 06:58:39 +0000 (06:58 +0000)]
Avoid misdetection of overlapping partitions...
As version-0.90 superblock don't record the superblock
offset, it is possible for overlapping partitions,
or a partition that starts on a 64K boundary in the whole device
to result in mis-detection - one partition or device might
be detected where the other was intended.
To avoid this awkward possibility, we reject assembly attempts
which seem to have two devices that are different but have the
same version-0.90 superblock.
To avoid this problem altogether, switch to version-1 metadata.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 06:12:19 +0000 (06:12 +0000)]
Remove libssl dependancy and always use sha1.c code
Apparently there are license issues with openssl, so
just use sha1.c always. This means we can get rid of
SHA1.c
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 05:53:22 +0000 (05:53 +0000)]
Get --stop to honour --quiet
And as a side effect, if --quiet isn't given, stopped devices are
reported.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 05:37:47 +0000 (05:37 +0000)]
Yet another include file that someone doesn't like....
It seems byteswap.h isn't universally accepted, so
we do it by hand...
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 05:33:40 +0000 (05:33 +0000)]
Allow default metadata to be specified in mdadm.conf
CREATE metadata=1
in mdadm.conf will cause version-1 superblocks to be the default.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 01:53:13 +0000 (01:53 +0000)]
Don't try to create a new device when using --manage or --grow
... as that's just silly!
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 01:49:00 +0000 (01:49 +0000)]
Finally (I hope) get byteorder stuff portable.
/usr/include/endian.h and /usr/include/byteswap.h together
can give us almost what we need...
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 01:34:30 +0000 (01:34 +0000)]
Fix typo in mail message from --monitor
From: martin f krafft <madduck@madduck.net>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 2 Jun 2006 01:32:45 +0000 (01:32 +0000)]
Fix error code from examine properly..
The last one was wrong :-(
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 04:09:21 +0000 (04:09 +0000)]
Improve compiling for static binaries.
Have "#ifdef STATIC" in config.c, and the 'rmconf' target
was a mess.
Instead, create 'pwgr.c' with stub routines for those unavailable
when statically compiled, and include that in STATICOBJ
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 04:02:15 +0000 (04:02 +0000)]
Make add_dev for uclibc return something.
As it is now 'int', it needs to.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:18:22 +0000 (02:18 +0000)]
Add uninstall target to Makefile
From: "Dirk Jagdmann" <jagdmann@gmail.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:18:22 +0000 (02:18 +0000)]
New install-* targets for installing differently compiled mdadms
e.g. install-static to install the statically compiled version, etc.
From: "Dirk Jagdmann" <jagdmann@gmail.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:18:22 +0000 (02:18 +0000)]
Makefile improvements for static linking.
From: "Dirk Jagdmann" <jagdmann@gmail.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:06:32 +0000 (02:06 +0000)]
check return status of all write/fwrite functions as required by glibc 2.4
From: Luca Berra <bluca@vodka.it>
glibc 2.4 is pedantic on ignoring return values from fprintf, fwrite and
write, so now we check the rval and actually do something with it.
in the Grow.c case i only print a warning, since i don't think we can do
anything in case we fail invalidating those superblocks (is should never
happen, but then...)
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:06:32 +0000 (02:06 +0000)]
Uclibc has deprecated 'random' so use 'rand' instead.
glibc says of rand
Do not use this function in applications intended to be
portable when good randomness is needed.
Go figure...
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:06:32 +0000 (02:06 +0000)]
add CFLAGS to mdassemble build and fix a couple of non-returning functions
pass CFLAGS to mdassemble build, enabling -Wall -Werror showed some
issues also fixed by the patch.
From: Luca Berra <bluca@vodka.it>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:06:32 +0000 (02:06 +0000)]
This is to avoid gcc warnings when building with strict-aliasing optimization
fix for another srict-aliasing problem, you can typecast a reference to a
void pointer to anything, you cannot typecast a reference to a
struct.
From: Luca Berra <bluca@vodka.it>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Mon, 29 May 2006 02:06:32 +0000 (02:06 +0000)]
snprintf size should be at most the size of the buffer
From: Luca Berra <bluca@vodka.it>
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 07:12:48 +0000 (07:12 +0000)]
Release 2.5
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 04:28:33 +0000 (04:28 +0000)]
Make sure everything compiles...
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 04:16:07 +0000 (04:16 +0000)]
Support --auto-update-homehost
This can be used to bootstrape homehost tagging.
If no arrays are found that are tagged, we look for any array
and tag it.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 03:24:02 +0000 (03:24 +0000)]
Set default name for v1 array based on device name.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 03:11:57 +0000 (03:11 +0000)]
Support auto-assembling of stacked devices
and assorted bugfixes.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 02:32:15 +0000 (02:32 +0000)]
Never remove device nodes of active devices.
This is just too risky.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 02:29:12 +0000 (02:29 +0000)]
Improve names reported by --examine --brief
Instead of depending (too much) on what is in /dev,
we make names based on the content of the superblock.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 01:40:53 +0000 (01:40 +0000)]
Choose better devnumbers and tidy up some issues with finding names.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Fri, 26 May 2006 00:50:15 +0000 (00:50 +0000)]
Initial implementation of auto-assembly
This basically works, but needs various improvements and some tests.
Signed-off-by: Neil Brown <neilb@suse.de>
Neil Brown [Tue, 23 May 2006 06:46:40 +0000 (06:46 +0000)]
Allow autoassembly to choose it's own name for the array.
This cannot be used yet, but it is working towards auto-assembly.
When auto-assembling an array, we make a name in /dev/md/
giving a number (from the peferred minor) or name (from set-name).
Signed-off-by: Neil Brown <neilb@suse.de>