[thirdparty/mdadm.git] / ChangeLog
11a3e71d 1Changes Prior to this release
2 - Fix another bug in Assemble.c due to confusing 'i' with 'j'
3 - Minimal, untested, support for multipath
4 - re-write of argument parsing to have more coherent modes,
5 - add --query,-Q option
6 - Update mdadm.8 to reflect arg processing change and --query
7 - Change "long" to "unsigned long" for device sizes
8 - Handle "mailaddr" and "program" lines in config file for follow/scan mode.
9 - --follow --scan will exit if no program or mail found
10 - Add MAILADDR and PROGRAM to mdadm.conf-example
11 - Spell check man pages
12 - consistently use "component devices" instead of "subdevices"
13 - Make -Wall -Werror really work and fix lots of errors.
14 - --detail and --stop can have --scan which chooses devices from /proc/mdstat
15 - --monitor detects 20% changes in resync, failed spares,
16 disappearing arrays,
17 - --monitor --scan will automatically add any devices found in /proc/mdstat
18 - --monitor will move spares between arrays with same spare-group if necessary
19 - Documentation for Monitor Mode
20 - --query notes if the array containing the given device is active or not
21 - Finished md.4 man page.
23Changes Prior to 0.7.2 release
24 - mdadm.spec updates and ifdef BLKGETSIZE64 from Luca Berra --
25 - more mdadm.spec updates from Gregory Leblanc <>
26 - make directory for mdadm.conf configurable in Makefile
27 - Finished mdadm.conf.5. Removed details of conf file from
28 mdadm.8 leaving a reference to mdadm.conf.5.
29 - Fix bug in Assemble.c, thanks to Junaid Rizvi <>
30 - Get --assemble --force to make sure old major/minor numbers are
31 consistant, as md.c worries about this :-(
34Changes Prior to 0.7.1 release
35 - update mdadm.spec
36 - use BLKGETSIZE64 if available for array size
37 - give human readable as GiB/MiB and GB and MB, with 2 decimal point precision
38 - Only warn about size variation for raid1/4/5.
39 - Started md.4 man page
40 - Started mdadm.conf.5 man page
42Changes Prior to 0.7 release
44 - Fix makefile to install binary at /sbin and not /sbin/sbin
45 Also install man page.
46 - Add --zero-superblock based on --destroywithextremeprejudice
47 from Dale Stephenson <>
48 - change name to mdadm. It is palandromic, and much nicer to pronouce.
50Changes Prior to 0.6 release
9a9dab36 52 - Remove the limit on the number of device names that can be
53 given on the command line.
54 - Fix bug in --assemble --force where it would only update a
55 single superblock.
56 - Fix bogus printing of big numbers not being block devices
57 when given names of devices that don't exist.
58 - When --assemble --force, consider superblocks with an event
59 count that is 1 behind as out-of-date. Normally they are
60 considered up-to-date (as the kernel assumes this too).
61 - When marking drives as not-failed in the superblock,
62 we also mark them as ACTIVE and SYNC.
63 - Don't start arrays for which not all drives are available unless:
64 --scan which implies that all drives were found automatically
65 --run which means the user knows what they want
66 --force which means that we are fixing something broken
67 - Make sure all device numbers passed as 3rd arg of ioctl
68 are passed as unsigned lock, so that it works on SPARC
69 - If HOT_ADD_DISK failes for -a, then only try ADD_NEW_DISK
70 if we cannot read from the array, i.e. if the array is
71 not started yet.
72 - man page update
73 - Taught Examine to handle --scan. It examines all devices listed
74 on DEVICE lines in the config file.
75 - Added --brief (-b) flag for Examine and Detail to print out
76 and mdctl.conf compatible description with uuid=, level=,
77 disks= and - for Examine - devices=
78 --examine --brief collects all devices the make the one array and
79 list them as one entry.
80 - Added level= and disks= options to ARRAY lines in config files
81 so --brief output could be used as-is.
82 - Make parity style ({left,right}-{,a}symmetric) consistantly use -,
83 never _.
84 - Add "Array Size" to --detail output
85 - Change "Size" to "Device Size" and exclude from Detail of arrays
86 that do not have a consistent device size.
87 - Add Human readable MiB or GiB value on size lines of Detail and Examine
88 - --assemble --scan doesn't complain about active drives
89 - require number of spares given in -x to be listed.
90 - Made --build actually work.
91Changes Prior to 0.5 release
93 --assemble:
94 spare drives are handled properly.
96 --force can be used to recover from 2-drive failures on RAID5
97 If you belive that /dev/hda1 /dev/hdb1 /dev/hdc1 /dev/hdd1 should
98 make a raid5 array, but it has experienced multiple failures and
99 wont start, then
101 mdctl --assemble --force /dev/md0 /dev/hd[abcd]1
103 Should update the superblock on the newest failed drive and
104 restart the array in degraded mode. You should then remove the
105 remaining failed drive and re-add it (if you are happy that it
106 might work).
108 Ofcourse whenever you have a 2-drive failure, you have a risk
109 of corruption in data that hasn't be changed for a long time. So
110 this doesn't give you your array back all nice and happy, but it
111 does allow you to recover data that might not be corrupt.
113 More flexibility in identifying a RAID array in the mdctl.conf
114 e.g.
115 array /dev/md4 super-minor=4
117 assembles /dev/md4 from all devices found that have a raid
118 superblock that says the minor number of the array is 4.
119 If the blocks with the right minor number do not all have the
120 same UUID, an error is flags and no assembly happens.
122 array /dev/md3 devices=/dev/hd[abc]2
124 Assembles /dev/md3 drom /dev/hda2 /dev/hdb2 and/dev/hdc2. All
125 devices must exist and have raid superblock with the same uuid.
127 If two identity specifiers are used, only devices that match all
128 of them are considered, so
130 array /dev/md2 devices=/dev/hd?2 super-minor=2
132 will assemble /dev/md2 using all /dev/hd?2 devices which have a
133 raid superblock with minor number 2.
135 --create:
136 When listing devices for --create, the word "missing" can be
137 used to indicate that the respective slot does not have a
138 working drive currently. This is similar to the "failed-disk"
139 directive in mkraid/raidtab.
140 e.g.
141 mdctl --create --level=5 -raid-disks=4 --spare-disks=2
142 /dev/md0 /dev/sda /dev/sdb missing /dev/sdc /dev/sdd /dev/sde
144 will create a raid5 array with the third slot empty, and two
145 spares.
147 By default, raid5 arrays are created with the last slot empty
148 and drive listed for the last slot added as a spare. If a
149 "missing" slot is given, or if --force is given, then --create
150 does exactly what you ask and doesn't try to be clever.
153 --follow / --monitor:
155 This is a new mode. I couldn't stop my self from picking a name
156 starting with F (as current modes start A,B,C,D,E) but I
157 relented and provided an alternate name that is somewhat more
158 meaningful.
160 In this mode, mdctl does not exit, but runs continuously and
161 periodically polls all the md devices to see if they have had
162 any interested state change.
163 The changes that it currently notices are:
164 Fail - an active disc fails
165 FailSpare - a spare, that was presumably being build, fails
166 ActiveSpare - a spare becomes active, presumably after a rebuild.
168 Options:
169 --mail mailaddress - send Email on any Fail* event
170 --program program - run the program on any event.
171 Args are: eventname mddevice subdevice(if-known)
172 --delay seconds - change from the default 60second pause
173 between polls.
175 I plan to add functionality to this mode to allow sharing of
176 spare drives. If an array is marks "spare-group=fred", and it
177 has a failed drive and no spares, and if some other array is
178 also "spare-group=fred" and it has no failed drives, but does
179 have a spare drive that is big enough, the spare will be moved
180 to the first array.
182 I also have the idea of adding a --grow mode which will re-organise
183 the data on an N disk raid0/4/5 array to be on an N+M disk array.
184 I have no concrete plans for this though.
186 I got rid of the "v" in the archive file name, and include the
187 version number in the directory created by the archive.
189 There is now a man page and mdctl.spec (for rpm) thanks to
190 Danilo Godec <>.
192 Ofcourse, the man page is now out of date and despite being based on
193 the --help output, is not wholy correct. After I get --follow
194 working properly, I plan to revise the various documentation and/or
195 the code to make sure the two match.