[thirdparty/mdadm.git] / ChangeLog
1Changes Prior to this release
2 - mdadm.spec updates and ifdef BLKGETSIZE64 from Luca Berra --
3 - more mdadm.spec updates from Gregory Leblanc <>
4 - make directory for mdadm.conf configurable in Makefile
5 - Finished mdadm.conf.5. Removed details of conf file from
6 mdadm.8 leaving a reference to mdadm.conf.5.
7 - Fix bug in Assemble.c, thanks to Junaid Rizvi <>
8 - Get --assemble --force to make sure old major/minor numbers are
9 consistant, as md.c worries about this :-(
12Changes Prior to 0.7.1 release
13 - update mdadm.spec
14 - use BLKGETSIZE64 if available for array size
15 - give human readable as GiB/MiB and GB and MB, with 2 decimal point precision
16 - Only warn about size variation for raid1/4/5.
17 - Started md.4 man page
18 - Started mdadm.conf.5 man page
20Changes Prior to 0.7 release
22 - Fix makefile to install binary at /sbin and not /sbin/sbin
23 Also install man page.
24 - Add --zero-superblock based on --destroywithextremeprejudice
25 from Dale Stephenson <>
26 - change name to mdadm. It is palandromic, and much nicer to pronouce.
28Changes Prior to 0.6 release
9a9dab36 30 - Remove the limit on the number of device names that can be
31 given on the command line.
32 - Fix bug in --assemble --force where it would only update a
33 single superblock.
34 - Fix bogus printing of big numbers not being block devices
35 when given names of devices that don't exist.
36 - When --assemble --force, consider superblocks with an event
37 count that is 1 behind as out-of-date. Normally they are
38 considered up-to-date (as the kernel assumes this too).
39 - When marking drives as not-failed in the superblock,
40 we also mark them as ACTIVE and SYNC.
41 - Don't start arrays for which not all drives are available unless:
42 --scan which implies that all drives were found automatically
43 --run which means the user knows what they want
44 --force which means that we are fixing something broken
45 - Make sure all device numbers passed as 3rd arg of ioctl
46 are passed as unsigned lock, so that it works on SPARC
47 - If HOT_ADD_DISK failes for -a, then only try ADD_NEW_DISK
48 if we cannot read from the array, i.e. if the array is
49 not started yet.
50 - man page update
51 - Taught Examine to handle --scan. It examines all devices listed
52 on DEVICE lines in the config file.
53 - Added --brief (-b) flag for Examine and Detail to print out
54 and mdctl.conf compatible description with uuid=, level=,
55 disks= and - for Examine - devices=
56 --examine --brief collects all devices the make the one array and
57 list them as one entry.
58 - Added level= and disks= options to ARRAY lines in config files
59 so --brief output could be used as-is.
60 - Make parity style ({left,right}-{,a}symmetric) consistantly use -,
61 never _.
62 - Add "Array Size" to --detail output
63 - Change "Size" to "Device Size" and exclude from Detail of arrays
64 that do not have a consistent device size.
65 - Add Human readable MiB or GiB value on size lines of Detail and Examine
66 - --assemble --scan doesn't complain about active drives
67 - require number of spares given in -x to be listed.
68 - Made --build actually work.
69Changes Prior to 0.5 release
71 --assemble:
72 spare drives are handled properly.
74 --force can be used to recover from 2-drive failures on RAID5
75 If you belive that /dev/hda1 /dev/hdb1 /dev/hdc1 /dev/hdd1 should
76 make a raid5 array, but it has experienced multiple failures and
77 wont start, then
79 mdctl --assemble --force /dev/md0 /dev/hd[abcd]1
81 Should update the superblock on the newest failed drive and
82 restart the array in degraded mode. You should then remove the
83 remaining failed drive and re-add it (if you are happy that it
84 might work).
86 Ofcourse whenever you have a 2-drive failure, you have a risk
87 of corruption in data that hasn't be changed for a long time. So
88 this doesn't give you your array back all nice and happy, but it
89 does allow you to recover data that might not be corrupt.
91 More flexibility in identifying a RAID array in the mdctl.conf
92 e.g.
93 array /dev/md4 super-minor=4
95 assembles /dev/md4 from all devices found that have a raid
96 superblock that says the minor number of the array is 4.
97 If the blocks with the right minor number do not all have the
98 same UUID, an error is flags and no assembly happens.
100 array /dev/md3 devices=/dev/hd[abc]2
102 Assembles /dev/md3 drom /dev/hda2 /dev/hdb2 and/dev/hdc2. All
103 devices must exist and have raid superblock with the same uuid.
105 If two identity specifiers are used, only devices that match all
106 of them are considered, so
108 array /dev/md2 devices=/dev/hd?2 super-minor=2
110 will assemble /dev/md2 using all /dev/hd?2 devices which have a
111 raid superblock with minor number 2.
113 --create:
114 When listing devices for --create, the word "missing" can be
115 used to indicate that the respective slot does not have a
116 working drive currently. This is similar to the "failed-disk"
117 directive in mkraid/raidtab.
118 e.g.
119 mdctl --create --level=5 -raid-disks=4 --spare-disks=2
120 /dev/md0 /dev/sda /dev/sdb missing /dev/sdc /dev/sdd /dev/sde
122 will create a raid5 array with the third slot empty, and two
123 spares.
125 By default, raid5 arrays are created with the last slot empty
126 and drive listed for the last slot added as a spare. If a
127 "missing" slot is given, or if --force is given, then --create
128 does exactly what you ask and doesn't try to be clever.
131 --follow / --monitor:
133 This is a new mode. I couldn't stop my self from picking a name
134 starting with F (as current modes start A,B,C,D,E) but I
135 relented and provided an alternate name that is somewhat more
136 meaningful.
138 In this mode, mdctl does not exit, but runs continuously and
139 periodically polls all the md devices to see if they have had
140 any interested state change.
141 The changes that it currently notices are:
142 Fail - an active disc fails
143 FailSpare - a spare, that was presumably being build, fails
144 ActiveSpare - a spare becomes active, presumably after a rebuild.
146 Options:
147 --mail mailaddress - send Email on any Fail* event
148 --program program - run the program on any event.
149 Args are: eventname mddevice subdevice(if-known)
150 --delay seconds - change from the default 60second pause
151 between polls.
153 I plan to add functionality to this mode to allow sharing of
154 spare drives. If an array is marks "spare-group=fred", and it
155 has a failed drive and no spares, and if some other array is
156 also "spare-group=fred" and it has no failed drives, but does
157 have a spare drive that is big enough, the spare will be moved
158 to the first array.
160 I also have the idea of adding a --grow mode which will re-organise
161 the data on an N disk raid0/4/5 array to be on an N+M disk array.
162 I have no concrete plans for this though.
164 I got rid of the "v" in the archive file name, and include the
165 version number in the directory created by the archive.
167 There is now a man page and mdctl.spec (for rpm) thanks to
168 Danilo Godec <>.
170 Ofcourse, the man page is now out of date and despite being based on
171 the --help output, is not wholy correct. After I get --follow
172 working properly, I plan to revise the various documentation and/or
173 the code to make sure the two match.