]>
Commit | Line | Data |
---|---|---|
1 | Subject: ANNOUNCE: mdadm 3.2 - A tool for managing Soft RAID under Linux (DEVEL ONLY) | |
2 | ||
3 | I am pleased to announce the availability of | |
4 | mdadm version 3.2 | |
5 | ||
6 | It is available at the usual places: | |
7 | countrycode=xx. | |
8 | http://www.${countrycode}kernel.org/pub/linux/utils/raid/mdadm/ | |
9 | and via git at | |
10 | git://neil.brown.name/mdadm devel-3.2 | |
11 | http://neil.brown.name/git?p=mdadm | |
12 | ||
13 | This is a "Developers only" release. Please don't consider using it | |
14 | or making it available to others without reading the following. | |
15 | ||
16 | ||
17 | By far the most significant change in this release related to the | |
18 | management of reshaping arrays. This code has been substantially | |
19 | re-written so that it can work with 'externally managed metadata' - | |
20 | Intel's IMSM in particular. We now support level migration and | |
21 | OnLine Capacity Expansion on these arrays. | |
22 | ||
23 | However, while the code largely works it has not been tested | |
24 | exhaustively so there are likely to be problems. As the reshape code | |
25 | for native metadata arrays was changed as part of this rewrite these | |
26 | problems could also result in regressions for reshape of native | |
27 | metadata. | |
28 | ||
29 | It is partly to encourage greater testing that this release is being | |
30 | made. Any reports of problem - particular reproducible recipes for | |
31 | triggering the problems - will be gratefully received. | |
32 | ||
33 | It is hopped that a "3.2.1" release will be available in early March | |
34 | which will be a bugfix release over this and can be considered | |
35 | suitable for general use. | |
36 | ||
37 | Other changes of note: | |
38 | ||
39 | - Policy framework. | |
40 | Various policy statements can be made in the mdadm.conf to guide | |
41 | the behaviour of mdadm, particular with regards to how new devices | |
42 | are treated by "mdadm -I". | |
43 | Depending on the 'action' associated with a device (identified by | |
44 | its 'path') such need devices can be automatically re-added to and | |
45 | existing array that they previously fell out off, or automatically | |
46 | added as a spare if they appear to contain no data. | |
47 | ||
48 | - mdadm now has a limited understanding of partition tables. This | |
49 | allows the policy framework to make decisions about partitioned | |
50 | devices as well. | |
51 | ||
52 | - --incremental --remove can be told what --path the device was on, | |
53 | and this info will be recorded so that another device appearing at | |
54 | the same physical location can be preferentially added to the same | |
55 | array (provides the spare-same-slot action policy applied to the | |
56 | path). | |
57 | ||
58 | - A new flags "--invalid-backup" flag is available in --assemble | |
59 | mode. This can be used to re-assemble an array which was stopping | |
60 | in the middle of a reshape, and for which the 'backup file' is no | |
61 | longer available or is corrupted. The array may have some | |
62 | corruption in it at the point where reshape was up to, but at least | |
63 | the rest of the array will become available. | |
64 | ||
65 | ||
66 | - Various internal restructuring - more is needed. | |
67 | ||
68 | ||
69 | Any feed back and bug reports are always welcomed at: | |
70 | linux-raid@vger.kernel.org | |
71 | ||
72 | And please: don't use this in production - particularly not the | |
73 | --grow functionality. | |
74 | ||
75 | NeilBrown 1st February 2011 | |
76 | ||
77 |