1 |
On Wednesday 04 May 2011 13:08:34 Evgeny Bushkov wrote: |
2 |
> On 04.05.2011 11:54, Joost Roeleveld wrote: |
3 |
> > On Wednesday 04 May 2011 10:07:58 Evgeny Bushkov wrote: |
4 |
> >> On 04.05.2011 01:49, Florian Philipp wrote: |
5 |
> >>> Am 03.05.2011 19:54, schrieb Evgeny Bushkov: |
6 |
> >>>> Hi. |
7 |
> >>>> How can I find out which is the parity disk in a RAID-4 soft |
8 |
> >>>> array? I |
9 |
> >>>> couldn't find that in the mdadm manual. I know that RAID-4 |
10 |
> >>>> features a |
11 |
> >>>> dedicated parity disk that is usually the bottleneck of the array, |
12 |
> >>>> so |
13 |
> >>>> that disk must be as fast as possible. It seems useful to employ a |
14 |
> >>>> few |
15 |
> >>>> slow disks with a relatively fast disk in such a RAID-4 array. |
16 |
> >>>> |
17 |
> >>>> Best regards, |
18 |
> >>>> Bushkov E. |
19 |
> >>> |
20 |
> >>> You are seriously considering a RAID4? You know, there is a reason |
21 |
> >>> why |
22 |
> >>> it was superseded by RAID5. Given the way RAID4 operates, a first |
23 |
> >>> guess |
24 |
> >>> for finding the parity disk in a running array would be the one with |
25 |
> >>> the worst SMART data. It is the parity disk that dies the soonest. |
26 |
> >>> |
27 |
> >>> From looking at the source code it seems like the last specified |
28 |
> >>> disk is parity. Disclaimer: I'm no kernel hacker and I have only |
29 |
> >>> inspected the code, not tried to understand the whole MD subsystem. |
30 |
> >>> |
31 |
> >>> Regards, |
32 |
> >>> Florian Philipp |
33 |
> >> |
34 |
> >> Thank you for answering... The reason I consider RAID-4 is a few |
35 |
> >> sata/150 drives and a pair of sata II drives I've got. Let's look at |
36 |
> >> the problem from the other side: I can create RAID-0(from sata II |
37 |
> >> drives) and then add it to RAID-4 as the parity disk. It doesn't |
38 |
> >> bother |
39 |
> >> me if any disk from the RAID-0 fails, that wouldn't disrupt my RAID-4 |
40 |
> >> array. For example: |
41 |
> >> |
42 |
> >> mdadm --create /dev/md1 --level=4 -n 3 -c 128 /dev/sdb1 /dev/sdc1 |
43 |
> >> missing mdadm --create /dev/md2 --level=0 -n 2 -c 128 /dev/sda1 |
44 |
> >> /dev/sdd1 mdadm /dev/md1 --add /dev/md2 |
45 |
> >> |
46 |
> >> livecd ~ # cat /proc/mdstat |
47 |
> >> Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] |
48 |
> >> md2 : active raid0 sdd1[1] sda1[0] |
49 |
> >> |
50 |
> >> 20969472 blocks super 1.2 128k chunks |
51 |
> >> |
52 |
> >> md1 : active raid4 md2[3] sdc1[1] sdb1[0] |
53 |
> >> |
54 |
> >> 20969216 blocks super 1.2 level 4, 128k chunk, algorithm 0 |
55 |
> >> [3/2] [UU_] |
56 |
> >> |
57 |
> >> [========>............] recovery = 43.7% (4590464/10484608) |
58 |
> >> finish=1.4min speed=69615K/sec |
59 |
> >> |
60 |
> >> That configuration works well, but I'm not sure if md1 is the parity |
61 |
> >> disk here, that's why I asked. May be I'm wrong and RAID-5 is the only |
62 |
> >> worth array, I'm just trying to consider all pros and cons here. |
63 |
> >> |
64 |
> >> Best regards, |
65 |
> >> Bushkov E. |
66 |
> > |
67 |
> > I only use RAID-0 (when I want performance and don't care about the |
68 |
> > data), RAID-1 (for data I can't afford to loose) and RAID-5 (data I |
69 |
> > would like to keep). I have never bothered with RAID-4. |
70 |
> > |
71 |
> > What do you see in the "dmesg" after the mdadm commands? |
72 |
> > It might actually mention which is the parity disk in there. |
73 |
> > |
74 |
> > -- |
75 |
> > Joost |
76 |
> |
77 |
> There's nothing special in dmesg: |
78 |
> |
79 |
> md: bind<md2> |
80 |
> RAID conf printout: |
81 |
> --- level:4 rd:3 wd:2 |
82 |
> disk 0, o:1, dev:sdb1 |
83 |
> disk 1, o:1, dev:sdc1 |
84 |
> disk 2, o:1, dev:md2 |
85 |
> md: recovery of RAID array md1 |
86 |
> |
87 |
> I've run some tests with different chunk sizes, the fastest was |
88 |
> raid-10(4 disks), raid-5(3 disks) was closely after. Raid-4(4 disks) was |
89 |
> almost as fast as raid-5 so I don't see any sense to use it. |
90 |
> |
91 |
> Best regards, |
92 |
> Bushkov E. |
93 |
|
94 |
What's the result of: |
95 |
mdadm --misc --detail /dev/md1 |
96 |
? |
97 |
|
98 |
Not sure what info this command will provide with a RAID-4... |
99 |
|
100 |
-- |
101 |
Joost |