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11 count that is 1 behind as out-of-date.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