[thirdparty/mdadm.git] / ChangeLog
11a3e71d 1Changes Prior to this release
2 - Correct spelling of persist[ae]nce/persist[ae]nt.
3 - Change "disk" to "device" in options and config file
4 - convert array size to "long long" *before* shift-left in -D and -Q
6Changes Prior to 0.8.1 release
7 - Add "INSTALL" file.
8 - Fix some "i" variables that were not being set properly
9 - Initialise minsize and maxsize so that compilers don't complain.
10 - Tidy up Makefile and mdadm.spec installations
11 - Add "multipath" to documentation of valid levels
13Changes Prior to 0.8 release
14 - Fix another bug in Assemble.c due to confusing 'i' with 'j'
15 - Minimal, untested, support for multipath
16 - re-write of argument parsing to have more coherent modes,
17 - add --query,-Q option
18 - Update mdadm.8 to reflect arg processing change and --query
19 - Change "long" to "unsigned long" for device sizes
20 - Handle "mailaddr" and "program" lines in config file for follow/scan mode.
21 - --follow --scan will exit if no program or mail found
22 - Add MAILADDR and PROGRAM to mdadm.conf-example
23 - Spell check man pages
24 - consistently use "component devices" instead of "subdevices"
25 - Make -Wall -Werror really work and fix lots of errors.
26 - --detail and --stop can have --scan which chooses devices from /proc/mdstat
27 - --monitor detects 20% changes in resync, failed spares,
28 disappearing arrays,
29 - --monitor --scan will automatically add any devices found in /proc/mdstat
30 - --monitor will move spares between arrays with same spare-group if necessary
31 - Documentation for Monitor Mode
32 - --query notes if the array containing the given device is active or not
33 - Finished md.4 man page.
35Changes Prior to 0.7.2 release
36 - mdadm.spec updates and ifdef BLKGETSIZE64 from Luca Berra --
37 - more mdadm.spec updates from Gregory Leblanc <>
38 - make directory for mdadm.conf configurable in Makefile
39 - Finished mdadm.conf.5. Removed details of conf file from
40 mdadm.8 leaving a reference to mdadm.conf.5.
41 - Fix bug in Assemble.c, thanks to Junaid Rizvi <>
42 - Get --assemble --force to make sure old major/minor numbers are
43 consistant, as md.c worries about this :-(
46Changes Prior to 0.7.1 release
47 - update mdadm.spec
48 - use BLKGETSIZE64 if available for array size
49 - give human readable as GiB/MiB and GB and MB, with 2 decimal point precision
50 - Only warn about size variation for raid1/4/5.
51 - Started md.4 man page
52 - Started mdadm.conf.5 man page
54Changes Prior to 0.7 release
56 - Fix makefile to install binary at /sbin and not /sbin/sbin
57 Also install man page.
58 - Add --zero-superblock based on --destroywithextremeprejudice
59 from Dale Stephenson <>
60 - change name to mdadm. It is palandromic, and much nicer to pronouce.
62Changes Prior to 0.6 release
9a9dab36 64 - Remove the limit on the number of device names that can be
65 given on the command line.
66 - Fix bug in --assemble --force where it would only update a
67 single superblock.
68 - Fix bogus printing of big numbers not being block devices
69 when given names of devices that don't exist.
70 - When --assemble --force, consider superblocks with an event
71 count that is 1 behind as out-of-date. Normally they are
72 considered up-to-date (as the kernel assumes this too).
73 - When marking drives as not-failed in the superblock,
74 we also mark them as ACTIVE and SYNC.
75 - Don't start arrays for which not all drives are available unless:
76 --scan which implies that all drives were found automatically
77 --run which means the user knows what they want
78 --force which means that we are fixing something broken
79 - Make sure all device numbers passed as 3rd arg of ioctl
80 are passed as unsigned lock, so that it works on SPARC
81 - If HOT_ADD_DISK failes for -a, then only try ADD_NEW_DISK
82 if we cannot read from the array, i.e. if the array is
83 not started yet.
84 - man page update
85 - Taught Examine to handle --scan. It examines all devices listed
86 on DEVICE lines in the config file.
87 - Added --brief (-b) flag for Examine and Detail to print out
88 and mdctl.conf compatible description with uuid=, level=,
89 disks= and - for Examine - devices=
90 --examine --brief collects all devices the make the one array and
91 list them as one entry.
92 - Added level= and disks= options to ARRAY lines in config files
93 so --brief output could be used as-is.
94 - Make parity style ({left,right}-{,a}symmetric) consistantly use -,
95 never _.
96 - Add "Array Size" to --detail output
97 - Change "Size" to "Device Size" and exclude from Detail of arrays
98 that do not have a consistent device size.
99 - Add Human readable MiB or GiB value on size lines of Detail and Examine
100 - --assemble --scan doesn't complain about active drives
101 - require number of spares given in -x to be listed.
102 - Made --build actually work.
103Changes Prior to 0.5 release
105 --assemble:
106 spare drives are handled properly.
108 --force can be used to recover from 2-drive failures on RAID5
109 If you belive that /dev/hda1 /dev/hdb1 /dev/hdc1 /dev/hdd1 should
110 make a raid5 array, but it has experienced multiple failures and
111 wont start, then
113 mdctl --assemble --force /dev/md0 /dev/hd[abcd]1
115 Should update the superblock on the newest failed drive and
116 restart the array in degraded mode. You should then remove the
117 remaining failed drive and re-add it (if you are happy that it
118 might work).
120 Ofcourse whenever you have a 2-drive failure, you have a risk
121 of corruption in data that hasn't be changed for a long time. So
122 this doesn't give you your array back all nice and happy, but it
123 does allow you to recover data that might not be corrupt.
125 More flexibility in identifying a RAID array in the mdctl.conf
126 e.g.
127 array /dev/md4 super-minor=4
129 assembles /dev/md4 from all devices found that have a raid
130 superblock that says the minor number of the array is 4.
131 If the blocks with the right minor number do not all have the
132 same UUID, an error is flags and no assembly happens.
134 array /dev/md3 devices=/dev/hd[abc]2
136 Assembles /dev/md3 drom /dev/hda2 /dev/hdb2 and/dev/hdc2. All
137 devices must exist and have raid superblock with the same uuid.
139 If two identity specifiers are used, only devices that match all
140 of them are considered, so
142 array /dev/md2 devices=/dev/hd?2 super-minor=2
144 will assemble /dev/md2 using all /dev/hd?2 devices which have a
145 raid superblock with minor number 2.
147 --create:
148 When listing devices for --create, the word "missing" can be
149 used to indicate that the respective slot does not have a
150 working drive currently. This is similar to the "failed-disk"
151 directive in mkraid/raidtab.
152 e.g.
153 mdctl --create --level=5 -raid-disks=4 --spare-disks=2
154 /dev/md0 /dev/sda /dev/sdb missing /dev/sdc /dev/sdd /dev/sde
156 will create a raid5 array with the third slot empty, and two
157 spares.
159 By default, raid5 arrays are created with the last slot empty
160 and drive listed for the last slot added as a spare. If a
161 "missing" slot is given, or if --force is given, then --create
162 does exactly what you ask and doesn't try to be clever.
165 --follow / --monitor:
167 This is a new mode. I couldn't stop my self from picking a name
168 starting with F (as current modes start A,B,C,D,E) but I
169 relented and provided an alternate name that is somewhat more
170 meaningful.
172 In this mode, mdctl does not exit, but runs continuously and
173 periodically polls all the md devices to see if they have had
174 any interested state change.
175 The changes that it currently notices are:
176 Fail - an active disc fails
177 FailSpare - a spare, that was presumably being build, fails
178 ActiveSpare - a spare becomes active, presumably after a rebuild.
180 Options:
181 --mail mailaddress - send Email on any Fail* event
182 --program program - run the program on any event.
183 Args are: eventname mddevice subdevice(if-known)
184 --delay seconds - change from the default 60second pause
185 between polls.
187 I plan to add functionality to this mode to allow sharing of
188 spare drives. If an array is marks "spare-group=fred", and it
189 has a failed drive and no spares, and if some other array is
190 also "spare-group=fred" and it has no failed drives, but does
191 have a spare drive that is big enough, the spare will be moved
192 to the first array.
194 I also have the idea of adding a --grow mode which will re-organise
195 the data on an N disk raid0/4/5 array to be on an N+M disk array.
196 I have no concrete plans for this though.
198 I got rid of the "v" in the archive file name, and include the
199 version number in the directory created by the archive.
201 There is now a man page and mdctl.spec (for rpm) thanks to
202 Danilo Godec <>.
204 Ofcourse, the man page is now out of date and despite being based on
205 the --help output, is not wholy correct. After I get --follow
206 working properly, I plan to revise the various documentation and/or
207 the code to make sure the two match.