| 1 |
On 04.05.2011 13:38, Joost Roeleveld wrote: |
| 2 |
> On Wednesday 04 May 2011 13:08:34 Evgeny Bushkov wrote: |
| 3 |
>> On 04.05.2011 11:54, Joost Roeleveld wrote: |
| 4 |
>>> On Wednesday 04 May 2011 10:07:58 Evgeny Bushkov wrote: |
| 5 |
>>>> On 04.05.2011 01:49, Florian Philipp wrote: |
| 6 |
>>>>> Am 03.05.2011 19:54, schrieb Evgeny Bushkov: |
| 7 |
>>>>>> Hi. |
| 8 |
>>>>>> How can I find out which is the parity disk in a RAID-4 soft |
| 9 |
>>>>>> array? I |
| 10 |
>>>>>> couldn't find that in the mdadm manual. I know that RAID-4 |
| 11 |
>>>>>> features a |
| 12 |
>>>>>> dedicated parity disk that is usually the bottleneck of the array, |
| 13 |
>>>>>> so |
| 14 |
>>>>>> that disk must be as fast as possible. It seems useful to employ a |
| 15 |
>>>>>> few |
| 16 |
>>>>>> slow disks with a relatively fast disk in such a RAID-4 array. |
| 17 |
>>>>>> |
| 18 |
>>>>>> Best regards, |
| 19 |
>>>>>> Bushkov E. |
| 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 it was superseded by RAID5.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 |
>>>> Thank you for answering... The reason I consider RAID-4 is a few |
| 34 |
>>>> sata/150 drives and a pair of sata II drives I've got. Let's look at |
| 35 |
>>>> the problem from the other side: I can create RAID-0(from sata II |
| 36 |
>>>> drives) and then add it to RAID-4 as the parity disk. It doesn't |
| 37 |
>>>> bother |
| 38 |
>>>> me if any disk from the RAID-0 fails, that wouldn't disrupt my RAID-4 |
| 39 |
>>>> array. For example: |
| 40 |
>>>> |
| 41 |
>>>> mdadm --create /dev/md1 --level=4 -n 3 -c 128 /dev/sdb1 /dev/sdc1 |
| 42 |
>>>> missing mdadm --create /dev/md2 --level=0 -n 2 -c 128 /dev/sda1 |
| 43 |
>>>> /dev/sdd1 mdadm /dev/md1 --add /dev/md2 |
| 44 |
>>>> |
| 45 |
>>>> livecd ~ # cat /proc/mdstat |
| 46 |
>>>> Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] |
| 47 |
>>>> md2 : active raid0 sdd1[1] sda1[0] |
| 48 |
>>>> |
| 49 |
>>>> 20969472 blocks super 1.2 128k chunks |
| 50 |
>>>> |
| 51 |
>>>> md1 : active raid4 md2[3] sdc1[1] sdb1[0] |
| 52 |
>>>> |
| 53 |
>>>> 20969216 blocks super 1.2 level 4, 128k chunk, algorithm 0 |
| 54 |
>>>> [3/2] [UU_] |
| 55 |
>>>> |
| 56 |
>>>> [========>............] recovery = 43.7% (4590464/10484608) |
| 57 |
>>>> finish=1.4min speed=69615K/sec |
| 58 |
>>>> |
| 59 |
>>>> That configuration works well, but I'm not sure if md1 is the parity |
| 60 |
>>>> disk here, that's why I asked. May be I'm wrong and RAID-5 is the only |
| 61 |
>>>> worth array, I'm just trying to consider all pros and cons here. |
| 62 |
>>>> |
| 63 |
>>>> Best regards, |
| 64 |
>>>> Bushkov E. |
| 65 |
>>> I only use RAID-0 (when I want performance and don't care about the |
| 66 |
>>> data), RAID-1 (for data I can't afford to loose) and RAID-5 (data I |
| 67 |
>>> would like to keep). I have never bothered with RAID-4. |
| 68 |
>>> |
| 69 |
>>> What do you see in the "dmesg" after the mdadm commands? |
| 70 |
>>> It might actually mention which is the parity disk in there. |
| 71 |
>>> |
| 72 |
>>> -- |
| 73 |
>>> Joost |
| 74 |
>> There's nothing special in dmesg: |
| 75 |
>> |
| 76 |
>> md: bind<md2> |
| 77 |
>> RAID conf printout: |
| 78 |
>> --- level:4 rd:3 wd:2 |
| 79 |
>> disk 0, o:1, dev:sdb1 |
| 80 |
>> disk 1, o:1, dev:sdc1 |
| 81 |
>> disk 2, o:1, dev:md2 |
| 82 |
>> md: recovery of RAID array md1 |
| 83 |
>> |
| 84 |
>> I've run some tests with different chunk sizes, the fastest was |
| 85 |
>> raid-10(4 disks), raid-5(3 disks) was closely after. Raid-4(4 disks) was |
| 86 |
>> almost as fast as raid-5 so I don't see any sense to use it. |
| 87 |
>> |
| 88 |
>> Best regards, |
| 89 |
>> Bushkov E. |
| 90 |
> What's the result of: |
| 91 |
> mdadm --misc --detail /dev/md1 |
| 92 |
> ? |
| 93 |
> |
| 94 |
> Not sure what info this command will provide with a RAID-4... |
| 95 |
> |
| 96 |
> -- |
| 97 |
> Joostlivecd ~ # mdadm --misc --detail /dev/md1 |
| 98 |
livecd ~ # mdadm --misc --detail /dev/md1 |
| 99 |
/dev/md1: |
| 100 |
Version : 1.2 |
| 101 |
Creation Time : Wed May 4 13:54:33 2011 |
| 102 |
Raid Level : raid4 |
| 103 |
Array Size : 122624 (119.77 MiB 125.57 MB) |
| 104 |
Used Dev Size : 61312 (59.89 MiB 62.78 MB) |
| 105 |
Raid Devices : 3 |
| 106 |
Total Devices : 3 |
| 107 |
Persistence : Superblock is persistent |
| 108 |
|
| 109 |
Update Time : Wed May 4 13:55:14 2011 |
| 110 |
State : clean |
| 111 |
Active Devices : 3 |
| 112 |
Working Devices : 3 |
| 113 |
Failed Devices : 0 |
| 114 |
Spare Devices : 0 |
| 115 |
|
| 116 |
Chunk Size : 128K |
| 117 |
|
| 118 |
Name : livecd:1 (local to host livecd) |
| 119 |
UUID : 654218f0:9f0b88d5:d82f39bc:ae08aa1e |
| 120 |
Events : 19 |
| 121 |
|
| 122 |
Number Major Minor RaidDevice State |
| 123 |
0 8 17 0 active sync /dev/sdb1 |
| 124 |
1 8 33 1 active sync /dev/sdc1 |
| 125 |
3 9 2 2 active sync /dev/md2 |
| 126 |
|
| 127 |
Best regards, |
| 128 |
Bushkov E. |
Archives