blob: 34755a6277281e7c2905c2aee78dbdef8829d3b0 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
|
metadata:
name: software-raid1
format: "Manual Test Definition 1.0"
description: "Use Linux utility mdadm to create, rebuild and delete
software RAID1. RAID1 consists of data mirroring, without
parity or striping."
maintainer:
- chase.qi@linaro.org
os:
- debian
- ubuntu
- centos
- fedora
scope:
- functional
devices:
- d02
- d03
- d05
- overdrive
- moonshot
- thunderX
environment:
- manual-test
run:
steps:
- Install OS on the SUT(system under test) and make sure it boots.
- Power off the SUT and install three extra hard drives(use sd(b|c|d)
here). The three hard drives shoud have the same mode, at least the
same capacity.
- Boot to OS and make sure mdadm utility installed.
- Create 'Linux RAID auto' partition on each of the three hard drives
by running the following steps.
- 1) "fdisk /dev/sdx"
- 2) Delete all existing partitions with fdisk command "d"
- 3) Create Linux raid auto partition with fdisk commands
"n -> p -> 1 -> enter -> enter -> t -> fd -> w"
- Run the following steps to test RAID1.
- 1) "mdadm --create /dev/md0 --level=mirror --raid-devices=2 /dev/sd[b-c]1"
- 2) Monitor '/proc/mdstat' to check if md0 created and running.
- 3) "mkfs.ext4 /dev/md0"
- 4) "mount /dev/md0 /mnt"
- 5) "echo 'RAID1 test' > /mnt raid1-test.txt"
- 6) Intentionally set faulty partition with command
"mdadm --manage --set-faulty /dev/md0 /dev/sdc1"
- 7) Execute "mdadm --detail /dev/md0" and check if RAID array shown as
'degraded' and sdc1 shown as 'faulty spare'.
- 8) Execute "mdadm --manage /dev/md0 -r /dev/sdc1" to remove sdc1.
- 9) Verify that '/mnt/raid1-test.txt' is not damaged.
- 10) Execute "mdadm --manage /dev/md0 -a /dev/sdd1" to add sdd1.
- 11) Monitor the output of "mdadm --detail /dev/md0" and make sure
that md0 'rebuilding' can be finished.
- Remove the md0 by running the following steps.
- 1) "umount /dev/md0"
- 2) "mdadm --stop /dev/md0"
- 3) "mdadm --remove /dev/md0"
- 4) "mdadm --zero-superblock /dev/sdb1 /dev/sdc1 /dev/sdd1"
expected:
- RAID1 array creating, deleting and rebuilding are successful.
- When one disk of RAID1 array damaged or removed, no data loss.
- When faulty disk replaced with a new one, RAID1 array can be rebuilt.
|