-
Notifications
You must be signed in to change notification settings - Fork 5.9k
/
rbd_mirror_stress.sh
executable file
·149 lines (121 loc) · 3.92 KB
/
rbd_mirror_stress.sh
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
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
#!/bin/sh
#
# rbd_mirror_stress.sh - stress test rbd-mirror daemon
#
IMAGE_COUNT=50
export LOCKDEP=0
if [ -n "${CEPH_REF}" ]; then
wget -O rbd_mirror_helpers.sh "https://git.ceph.com/?p=ceph.git;a=blob_plain;hb=$CEPH_REF;f=qa/workunits/rbd/rbd_mirror_helpers.sh"
. ./rbd_mirror_helpers.sh
else
. $(dirname $0)/rbd_mirror_helpers.sh
fi
create_snap()
{
local cluster=$1
local pool=$2
local image=$3
local snap_name=$4
rbd --cluster ${cluster} -p ${pool} snap create ${image}@${snap_name} \
--debug-rbd=20 --debug-journaler=20 2> ${TEMPDIR}/rbd-snap-create.log
}
compare_image_snaps()
{
local pool=$1
local image=$2
local snap_name=$3
local rmt_export=${TEMPDIR}/${CLUSTER2}-${pool}-${image}.export
local loc_export=${TEMPDIR}/${CLUSTER1}-${pool}-${image}.export
rm -f ${rmt_export} ${loc_export}
rbd --cluster ${CLUSTER2} -p ${pool} export ${image}@${snap_name} ${rmt_export}
rbd --cluster ${CLUSTER1} -p ${pool} export ${image}@${snap_name} ${loc_export}
cmp ${rmt_export} ${loc_export}
}
wait_for_pool_healthy()
{
local cluster=$1
local pool=$2
local image_count=$3
local s
local count
local state
for s in `seq 1 40`; do
sleep 30
count=$(rbd --cluster ${cluster} -p ${pool} mirror pool status | grep 'images: ')
test "${count}" = "images: ${image_count} total" || continue
state=$(rbd --cluster ${cluster} -p ${pool} mirror pool status | grep 'health:')
test "${state}" = "health: ERROR" && return 1
test "${state}" = "health: OK" && return 0
done
return 1
}
start_mirror ${CLUSTER1}
start_mirror ${CLUSTER2}
testlog "TEST: add image and test replay after client crashes"
image=test
create_image ${CLUSTER2} ${POOL} ${image} '512M'
wait_for_image_replay_started ${CLUSTER1} ${POOL} ${image}
for i in `seq 1 10`
do
stress_write_image ${CLUSTER2} ${POOL} ${image}
test_status_in_pool_dir ${CLUSTER1} ${POOL} ${image} 'up+replaying' 'master_position'
snap_name="snap${i}"
create_snap ${CLUSTER2} ${POOL} ${image} ${snap_name}
wait_for_snap_present ${CLUSTER1} ${POOL} ${image} ${snap_name}
compare_image_snaps ${POOL} ${image} ${snap_name}
done
for i in `seq 1 10`
do
snap_name="snap${i}"
remove_snapshot ${CLUSTER2} ${POOL} ${image} ${snap_name}
done
remove_image ${CLUSTER2} ${POOL} ${image}
wait_for_image_present ${CLUSTER1} ${POOL} ${image} 'deleted'
testlog "TEST: create many images"
for i in `seq 1 ${IMAGE_COUNT}`
do
image="image_${i}"
create_image ${CLUSTER2} ${POOL} ${image} '128M'
write_image ${CLUSTER2} ${POOL} ${image} 100
done
wait_for_pool_healthy ${CLUSTER2} ${POOL} ${IMAGE_COUNT}
wait_for_pool_healthy ${CLUSTER1} ${POOL} ${IMAGE_COUNT}
testlog "TEST: compare many images"
for i in `seq 1 ${IMAGE_COUNT}`
do
image="image_${i}"
wait_for_image_replay_started ${CLUSTER1} ${POOL} ${image}
wait_for_replay_complete ${CLUSTER1} ${CLUSTER2} ${POOL} ${image}
compare_images ${POOL} ${image}
done
testlog "TEST: delete many images"
for i in `seq 1 ${IMAGE_COUNT}`
do
image="image_${i}"
remove_image ${CLUSTER2} ${POOL} ${image}
done
testlog "TEST: image deletions should propagate"
wait_for_pool_healthy ${CLUSTER1} ${POOL} 0
for i in `seq 1 ${IMAGE_COUNT}`
do
image="image_${i}"
wait_for_image_present ${CLUSTER1} ${POOL} ${image} 'deleted'
done
testlog "TEST: delete images during bootstrap"
set_pool_mirror_mode ${CLUSTER1} ${POOL} 'image'
set_pool_mirror_mode ${CLUSTER2} ${POOL} 'image'
start_mirror ${CLUSTER1}
image=test
for i in `seq 1 10`
do
image="image_${i}"
create_image ${CLUSTER2} ${POOL} ${image} '512M'
enable_mirror ${CLUSTER2} ${POOL} ${image}
stress_write_image ${CLUSTER2} ${POOL} ${image}
wait_for_image_present ${CLUSTER1} ${POOL} ${image} 'present'
disable_mirror ${CLUSTER2} ${POOL} ${image}
wait_for_image_present ${CLUSTER1} ${POOL} ${image} 'deleted'
purge_snapshots ${CLUSTER2} ${POOL} ${image}
remove_image_retry ${CLUSTER2} ${POOL} ${image}
done
echo OK