int fd;
memset(&inf, 0, sizeof(inf));
- fd = open(devlist->devname, O_RDONLY, 0);
+ fd = open(devlist->devname, O_RDONLY);
if (fd >= 0 &&
ioctl(fd, GET_ARRAY_INFO, &inf) == 0 &&
inf.raid_disks == 0) {
minsize = freesize;
}
if (runstop != 1 || verbose >= 0) {
- int fd = open(dname, O_RDONLY, 0);
+ int fd = open(dname, O_RDONLY);
if (fd <0 ) {
fprintf(stderr, Name ": Cannot open %s: %s\n",
dname, strerror(errno));
inf->disk.state |= (1<<MD_DISK_WRITEMOSTLY);
if (st->ss->external && st->subarray[0])
- fd = open(dv->devname, O_RDWR, 0);
+ fd = open(dv->devname, O_RDWR);
else
- fd = open(dv->devname, O_RDWR|O_EXCL,0);
+ fd = open(dv->devname, O_RDWR|O_EXCL);
if (fd < 0) {
fprintf(stderr, Name ": failed to open %s "
* GET_ARRAY_INFO and GET_DISK_INFO ioctl calls
*/
- int fd = open(dev, O_RDONLY, 0);
+ int fd = open(dev, O_RDONLY);
int vers;
mdu_array_info_t array;
mdu_disk_info_t *disks;
}
if (brief) {
+ mdu_bitmap_file_t bmf;
printf("ARRAY %s level=%s num-devices=%d", dev,
c?c:"-unknown-",
array.raid_disks );
else
printf(" metadata=%02d.%02d",
array.major_version, array.minor_version);
+
+ /* Only try GET_BITMAP_FILE for 0.90.01 and later */
+ if (vers >= 9001 &&
+ ioctl(fd, GET_BITMAP_FILE, &bmf) == 0 &&
+ bmf.pathname[0]) {
+ printf(" bitmap=%s", bmf.pathname);
+ }
} else {
mdu_bitmap_file_t bmf;
unsigned long long larray_size;
last_block = nstripe * ndata;
ostripe = last_block / odata / (ochunk/512) * (ochunk/512);
}
- printf("mdadm: Need to backup %lluK of critical section..\n", last_block/2);
+ fprintf(stderr, Name ": Need to backup %lluK of critical "
+ "section..\n", last_block/2);
sra = sysfs_read(fd, 0,
GET_COMPONENT|GET_DEVS|GET_OFFSET|GET_STATE|
}
spares = sra->array.spare_disks;
if (backup_file) {
- fdlist[d] = open(backup_file, O_RDWR|O_CREAT|O_EXCL, 0600);
+ fdlist[d] = open(backup_file, O_RDWR|O_CREAT|O_EXCL, S_IRUSR | S_IWUSR);
if (fdlist[d] < 0) {
fprintf(stderr, Name ": %s: cannot create backup file %s: %s\n",
devname, backup_file, strerror(errno));
if (backup_file)
unlink(backup_file);
- printf(Name ": ... critical section passed.\n");
+ fprintf(stderr, Name ": ... critical section passed.\n");
break;
}
return 0;
TAGS :
etags *.h *.c
+
+DISTRO_MAKEFILE := $(wildcard distropkg/Makefile)
+ifdef DISTRO_MAKEFILE
+include $(DISTRO_MAKEFILE)
+endif
+
sysfs_free(mdi);
if (fd >= 0 && ioctl(fd, STOP_ARRAY, NULL)) {
- if (quiet == 0)
+ if (quiet == 0) {
fprintf(stderr, Name
": failed to stop array %s: %s\n",
devname, strerror(errno));
+ if (errno == EBUSY)
+ fprintf(stderr, "Perhaps a running "
+ "process, mounted filesystem "
+ "or active volume group?\n");
+ }
return 1;
}
return 1;
}
close(0);
- open("/dev/null", 3);
+ open("/dev/null", O_RDWR);
dup2(0,1);
dup2(0,2);
setsid();
* whether it is an md device and whether it has
* a superblock
*/
- int fd = open(dev, O_RDONLY, 0);
+ int fd = open(dev, O_RDONLY);
int vers;
int ioctlerr;
int superror, superrno;
" --size= -z : Change the active size of devices in an array.\n"
" : This is useful if all devices have been replaced\n"
" : with larger devices.\n"
-" --raid-disks= -n : Change the number of active devices in an array.\n"
-" : array.\n"
+" --raid-devices= -n : Change the number of active devices in an array.\n"
" --bitmap= -b : Add or remove a write-intent bitmap.\n"
+" --backup-file= file : A file on a differt device to store data for a\n"
+" : short time while increasing raid-devices on a\n"
+" : RAID4/5/6 array. Not needed when a spare is present.\n"
;
char Help_incr[] =
of any given block are on different drives.
The 'far' arrangement can give sequential read performance equal to
-that of a RAID0 array, but at the cost of degraded write performance.
+that of a RAID0 array, but at the cost of reduced write performance.
When 'offset' replicas are chosen, the multiple copies of a given
chunk are laid out on consecutive drives and at consecutive offsets.
is also not true RAID, and it only involves one device. It
provides a layer over a true device that can be used to inject faults.
-.\".B mdadm
+.\".I mdadm
.\"is a program that can be used to create, manage, and monitor
.\"MD devices. As
.\"such it provides a similar set of functionality to the
.\".B raidtools
.\"packages.
.\"The key differences between
-.\".B mdadm
+.\".I mdadm
.\"and
.\".B raidtools
.\"are:
.\".IP \(bu 4
-.\".B mdadm
+.\".I mdadm
.\"is a single program and not a collection of programs.
.\".IP \(bu 4
-.\".B mdadm
+.\".I mdadm
.\"can perform (almost) all of its functions without having a
.\"configuration file and does not use one by default. Also
-.\".B mdadm
+.\".I mdadm
.\"helps with management of the configuration
.\"file.
.\".IP \(bu 4
-.\".B mdadm
+.\".I mdadm
.\"can provide information about your arrays (through Query, Detail, and Examine)
.\"that
.\".B raidtools
Assemble the components of a previously created
array into an active array. Components can be explicitly given
or can be searched for.
-.B mdadm
+.I mdadm
checks that the components
do form a bona fide array, and can, on request, fiddle superblock
information so as to assemble a faulty array.
.TP
.BR \-q ", " \-\-quiet
Avoid printing purely informative messages. With this,
-.B mdadm
+.I mdadm
will be silent unless there is something really important to report.
.TP
.B /proc/mdstat
for missing information.
In general, this option gives
-.B mdadm
+.I mdadm
permission to get any missing information (like component devices,
array devices, array identities, and alert destination) from the
configuration file (see previous option);
to use the minor number of the md device that is being assembled.
e.g. when assembling
.BR /dev/md0 ,
-.M \-\-super\-minor=dev
+.B \-\-super\-minor=dev
will look for super blocks with a minor number of 0.
.TP
.TP
.BR \-d ", " \-\-delay
Give a delay in seconds.
-.B mdadm
+.I mdadm
polls the md arrays and then waits this many seconds before polling
again. The default is 60 seconds.
.TP
.BR \-f ", " \-\-daemonise
Tell
-.B mdadm
+.I mdadm
to run as a background daemon if it decides to monitor anything. This
causes it to fork and run in the child, and to disconnect form the
terminal. The process id of the child is written to stdout.
.TP
.BR \-i ", " \-\-pid\-file
When
-.B mdadm
+.I mdadm
is running in daemon mode, write the pid of the daemon process to
the specified file, instead of printing it on standard output.
To create a "degraded" array in which some devices are missing, simply
give the word "\fBmissing\fP"
in place of a device name. This will cause
-.B mdadm
+.I mdadm
to leave the corresponding slot in the array empty.
For a RAID4 or RAID5 array at most one slot can be
"\fBmissing\fP"; for a RAID6 array at most two slots.
"\fBmissing\fP".
When creating a RAID5 array,
-.B mdadm
+.I mdadm
will automatically create a degraded array with an extra spare drive.
This is because building the spare into a degraded array is in general faster than resyncing
the parity on a non-degraded, but not clean, array. This feature can
.B \-\-force
option.
-When creating an array with version-1 metadata a name for the host is
+When creating an array with version-1 metadata a name for the array is
required.
If this is not given with the
.B \-\-name
option,
.I mdadm
-will chose a name based on the last component of the name of the
+will choose a name based on the last component of the name of the
device being created. So if
.B /dev/md3
is being created, then the name
.B home
will be used.
+When creating a partition based array, using
+.I mdadm
+with version-1.x metadata, the partition type should be set to
+.B 0xDA
+(non fs-data). This type selection allows for greater precision since
+using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
+might create problems in the event of array recovery through a live cdrom.
+
A new array will normally get a randomly assigned 128bit UUID which is
very likely to be unique. If you have a specific need, you can choose
a UUID for the array by giving the
.TP
.B \-\-examine
The device should be a component of an md array.
-.B mdadm
+.I mdadm
will read the md superblock of the device and display the contents.
If
.B \-\-brief
.PP
This usage causes
-.B mdadm
+.I mdadm
to periodically poll a number of md arrays and to report on any events
noticed.
-.B mdadm
+.I mdadm
will never exit once it decides that there are arrays to be checked,
so it should normally be run in the background.
As well as reporting events,
-.B mdadm
+.I mdadm
may move a spare drive from one array to another if they are in the
same
.B spare-group
and if the destination array has a failed drive but no spares.
If any devices are listed on the command line,
-.B mdadm
+.I mdadm
will only monitor those devices. Otherwise all arrays listed in the
configuration file will be monitored. Further, if
.B \-\-scan
.B \-\-scan
is given, then a program or an E-mail address must be specified on the
command line or in the config file. If neither are available, then
-.B mdadm
+.I mdadm
will not monitor anything.
Without
.B \-\-scan,
-.B mdadm
+.I mdadm
will continue monitoring as long as something was found to monitor. If
no program or email is given, then each event is reported to
.BR stdout .
the second device is the array that the spare was moved from.
For
-.B mdadm
+.I mdadm
to move spares from one array to another, the different arrays need to
be labeled with the same
.B spare-group
groups use different names.
When
-.B mdadm
+.I mdadm
detects that an array in a spare group has fewer active
devices than necessary for the complete array, and has no spare
devices, it will look for another array in the same spare group that
change the "size" attribute
for RAID1, RAID5 and RAID6.
.IP \(bu 4
-increase the "raid-disks" attribute of RAID1, RAID5, and RAID6.
+increase the "raid\-devices" attribute of RAID1, RAID5, and RAID6.
.IP \(bu 4
add a write-intent bitmap to any array which supports these bitmaps, or
remove a write-intent bitmap from such an array.
As an alternative,
.B \-\-run
may be passed to
-.B mdadm
+.I mdadm
in which case the array will be run as soon as there are enough
devices present for the data to be accessible. For a raid1, that
means one device will start the array. For a clean raid5, the array
filesystem,
.B /proc/mdstat
lists all active md devices with information about them.
-.B mdadm
+.I mdadm
uses this to find arrays when
.B \-\-scan
is given in Misc mode, and to monitor array reconstruction
Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
.SH NOTE
-.B mdadm
+.I mdadm
was previously known as
-.BR mdctl .
+.IR mdctl .
.P
-.B mdadm
+.I mdadm
is completely separate from the
-.B raidtools
+.I raidtools
package, and does not use the
.I /etc/raidtab
configuration file at all.
break;
}
if (raiddisks == 0) {
- fprintf(stderr, Name ": no raid-disks specified.\n");
+ fprintf(stderr, Name ": no raid-devices specified.\n");
rv = 1;
break;
}
break;
}
if (raiddisks == 0) {
- fprintf(stderr, Name ": no raid-disks specified.\n");
+ fprintf(stderr, Name ": no raid-devices specified.\n");
rv = 1;
break;
}
/etc/mdadm.conf
.SH DESCRIPTION
.PP
-.B mdadm
+.I mdadm
is a tool for creating, managing, and monitoring RAID devices using the
.B md
driver in Linux.
line lists the devices (whole devices or partitions) that might contain
a component of an MD array. When looking for the components of an
array,
-.B mdadm
+.I mdadm
will scan these devices (or any devices listed on the command line).
The
.B spare\-group
name are considered to be part of the same group. The significance of
a group of arrays is that
-.B mdadm
+.I mdadm
will, when monitoring the arrays, move a spare drive from one array in
a group to another array in that group if the first array had a failed
or missing drive but no spare.
.TP
.B auto=
This option declares to
-.B mdadm
+.I mdadm
that it should try to create the device file of the array if it
doesn't already exist, or exists but with the wrong device number.
.B mailaddr
line gives an E-mail address that alerts should be
sent to when
-.M mdadm
+.I mdadm
is running in
.B \-\-monitor
mode (and was given the
to suppress this symlink creation.
.RE
+.TP
+.B HOMEHOST
+The
+.B homehost
+line gives a default value for the
+.B --homehost=
+option to mdadm. There should be exactly one other word on the line.
+It should either exactly
+.B <system>
+or a host name.
+If
+.B <system>
+is given, then the
+.BR gethostname ( 2 )
+systemcall is used to get the host name.
+When arrays are created, this host name will be stored in the
+metadata. When arrays are assembled using auto-assembly, only arrays
+with this host name stored in the metadata will be considered.
.SH EXAMPLE
DEVICE /dev/sd[bcdjkl]1
/* from mdopen.c */
int open_mddev(char *dev, int autof/*unused */)
{
- int mdfd = open(dev, O_RDWR, 0);
+ int mdfd = open(dev, O_RDWR);
if (mdfd < 0)
fprintf(stderr, Name ": error opening %s: %s\n",
dev, strerror(errno));
must_remove = 1;
if (stb.st_mode && !must_remove) {
/* looks ok, see if it is available */
- mdfd = open(dev, O_RDWR, 0);
+ mdfd = open(dev, O_RDWR);
if (mdfd < 0) {
fprintf(stderr, Name ": error opening %s: %s\n",
dev, strerror(errno));
make_parts(dev,parts, ci->symlinks);
}
}
- mdfd = open(dev, O_RDWR, 0);
+ mdfd = open(dev, O_RDWR);
if (mdfd < 0)
fprintf(stderr, Name ": error opening %s: %s\n",
dev, strerror(errno));
path0=$dev6
path1=$dev7
+echo 2000 > /proc/sys/dev/raid/speed_limit_max
+echo 0 > /sys/module/md_mod/parameters/start_ro
+
if [ " $1" = " setup" ]
then trap 0 ; exit 0
fi
grep -s "active $1 " /proc/mdstat > /dev/null || {
echo >&2 "ERROR active $1 not found" ; cat /proc/mdstat ; exit 1;}
;;
- resync | recovery )
+ resync | recovery | reshape)
sleep 0.5
grep -s $1 /proc/mdstat > /dev/null || {
echo >&2 ERROR no $1 happening; cat /proc/mdstat; exit 1; }
nosync )
sleep 0.5
- if grep -s 're[synccovery]* =' > /dev/null /proc/mdstat ; then
+ if grep -s -E '(resync|recovery|reshape) =' > /dev/null /proc/mdstat ; then
echo >&2 "ERROR resync or recovery is happening!"; cat /proc/mdstat ; exit 1;
fi
;;
wait )
sleep 0.1
- while grep 're[synccovery]* =' > /dev/null /proc/mdstat
+ while grep -E '(resync|recovery|reshape|check|repair) =' > /dev/null /proc/mdstat
do sleep 2;
done
;;
mdadm -As -c /dev/null --homehost=testing -vvv
testdev $md1 1 $mdsize0 64
testdev $md0 1 $[mdsize0+mdsize00] 64
+mdadm -Ss
--- /dev/null
+
+#
+# test interrupting and restartign raid5 reshape.
+set -x
+devs="$dev1"
+st=UU
+for disks in 2 3 4 5
+do
+ eval devs=\"$devs \$dev$disks\"
+ st=U$st
+ for d in $devs
+ do dd if=/dev/urandom of=$d bs=1024 || true
+ done
+
+ mdadm -CR $md0 -amd -l5 -n$disks --assume-clean $devs
+ mdadm $md0 --add $dev6
+ echo 20 > /proc/sys/dev/raid/speed_limit_max
+ mdadm --grow $md0 -n $[disks+1]
+ check reshape
+ check state $st
+ mdadm --stop $md0
+ mdadm --assemble $md0 $devs $dev6
+ check reshape
+ echo 2000 > /proc/sys/dev/raid/speed_limit_max
+ check wait
+ echo check > /sys/block/md0/md/sync_action
+ check wait
+ mm=`cat /sys/block/md0/md/mismatch_cnt`
+ if [ $mm -gt 0 ]
+ then echo >&2 "ERROR mismatch_cnt non-zero : $mm" ; exit 1
+ fi
+ mdadm -S $md0
+done
# test restore: make a raid5 from a file, then do a compare
dd if=/dev/urandom of=/tmp/RandFile bs=1024 count=$size
$dir/test_stripe restore /tmp/RandFile $disks $[chunk*1024] 5 $nlayout 0 $[size*1024] $devs
- $mdadm -CR $md0 -amd -l5 -n$disks --assume-clean -c $chunk -p $layout $devs
+ mdadm -CR $md0 -amd -l5 -n$disks --assume-clean -c $chunk -p $layout $devs
cmp -s -n $[size*1024] $md0 /tmp/RandFile || { echo cmp failed ; exit 2; }
# FIXME check parity
> /tmp/NewRand
$dir/test_stripe save /tmp/NewRand $disks $[chunk*1024] 5 $nlayout 0 $[size*1024] $devs
cmp -s -n $[size*1024] $md0 /tmp/NewRand || { echo cmp failed ; exit 2; }
- $mdadm -S $md0
+ mdadm -S $md0
done
done
done
projects / thirdparty / mdadm.git / commitdiff
