-
Notifications
You must be signed in to change notification settings - Fork 26
/
Copy pathremotecords.py
executable file
·244 lines (198 loc) · 8.93 KB
/
remotecords.py
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
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
#! /usr/bin/env python
# Copyright (c) 2016 Aishwarya Ganesan and Ramnatthan Alagappan.
# All Rights Reserved.
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import os
import sys
import math
from collections import defaultdict
import subprocess
import argparse
import time
from threading import Timer
BLOCKSIZE = 4096
remote_user_name = 'aishwarya'
ERRFS_HOME = os.path.dirname(os.path.realpath(__file__))
fuse_command_err = 'nohup ' + ERRFS_HOME + "/errfs -f -omodules=subdir,subdir=%s %s err %s %s %s > /dev/null 2>&1 &"
fuse_unmount_command = "fusermount -u %s > /dev/null"
uppath = lambda _path, n: os.sep.join(_path.split(os.sep)[:-n])
parser = argparse.ArgumentParser()
parser.add_argument('--trace_files', nargs='+', required = True, help = 'Trace file paths')
parser.add_argument('--machine_ips', nargs='+', required = True, help = 'Machine ips')
parser.add_argument('--data_dirs', nargs='+', required = True, help = 'Location of data directories')
parser.add_argument('--workload_command', type = str)
parser.add_argument('--cords_results_base_dir', type = str, default='/run/shm', help = 'Location for checker state and results')
parser.add_argument('--checker_command', type = str, help = 'Checker for workloads that modify application state')
uppath = lambda _path, n: os.sep.join(_path.split(os.sep)[:-n])
def kill_proc(proc, timeout):
timeout["value"] = True
proc.kill()
def invoke_cmd(cmd):
p = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
out, err = p.communicate()
return (out, err)
def invoke_remote_cmd(machine_ip, command):
cmd = 'ssh {0}@{1} \'{2}\''.format(remote_user_name, machine_ip, command)
p = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
out, err = p.communicate()
return (out, err)
def copy_dir_from_remote(machine_ip, from_file_path, to_file_path):
cmd = 'scp -r {0}@{1}:{2} {3}'.format(remote_user_name, machine_ip, from_file_path, to_file_path)
os.system(cmd)
def block_roundup(x):
return math.ceil(x * 1.0 / BLOCKSIZE) * BLOCKSIZE
def block_rounddown(x):
return math.floor(x * 1.0 / BLOCKSIZE) * BLOCKSIZE
def get_block_nrs(offset, size):
start_offset = offset
end_offset = start_offset + size
total_blocks_touched = int((block_roundup(end_offset) - block_rounddown(start_offset)) / BLOCKSIZE)
assert total_blocks_touched >= 1
start_block_nr = int(math.floor(start_offset / BLOCKSIZE))
return range(start_block_nr, start_block_nr + total_blocks_touched)
data_dir_snapshots = []
data_dir_mount_points = []
args = parser.parse_args()
for i in range(0, len(args.trace_files)):
args.trace_files[i] = os.path.abspath(args.trace_files[i])
for i in range(0, len(args.data_dirs)):
args.data_dirs[i] = os.path.abspath(args.data_dirs[i])
data_dir_snapshots.append(os.path.join(uppath(args.data_dirs[i], 1), os.path.basename(os.path.normpath(args.data_dirs[i]))+ ".snapshot"))
data_dir_mount_points.append(os.path.join(uppath(args.data_dirs[i], 1), os.path.basename(os.path.normpath(args.data_dirs[i]))+ ".mp"))
trace_files = args.trace_files
data_dirs = args.data_dirs
assert len(trace_files) == len(data_dirs) and len(trace_files) == len(data_dir_snapshots) and len(trace_files) == len(data_dir_mount_points)
workload_command = args.workload_command
checker_command = args.checker_command
cords_results_base_dir = args.cords_results_base_dir
machine_ips = args.machine_ips
os.system('rm -rf ' + cords_results_base_dir+'/*')
replay_check_needed = False
if checker_command is not None and len(checker_command) > 0:
replay_check_needed = True
err_map = defaultdict(set)
machines = []
machine = 0
for trace_file in trace_files:
machines.append(machine)
assert os.path.exists(trace_file)
assert os.stat(trace_file).st_size > 0
with open(trace_file, "r") as f:
for line in f:
line = line.split('\t')
if line[0] in ['rename', 'unlink', 'link', 'symlink']:
pass
else:
assert len(line) == 4
filename = line[0]
op = line[1]
assert op == 'w' or op == 'r' or op == 'a'
offset = int(line[2])
size = int(line[3])
block_nrs = get_block_nrs(offset, size)
for block_nr in block_nrs:
err_map[(machine, filename)].add((op, block_nr))
machine += 1
assert len(machines) > 0
def get_error_modes(op):
if op == 'r':
return ["eio", "cz", "cg"]
elif op == 'w':
return ["eio"]
elif op == 'a':
return ["eio", "esp"]
else:
assert False
def cords_count():
total = 0
for key in err_map:
corrupt_machine = key[0]
other_machines = [i for i in machines if i != corrupt_machine]
assert len([corrupt_machine] + other_machines) == len(machines)
corrupt_filename = key[1]
block_ops = list(err_map[key])
for (op, block) in block_ops:
possible_err_modes = get_error_modes(op)
for err_type in possible_err_modes:
total += 1
return total
def cords_check():
total = cords_count()
count = 0
for key in err_map:
corrupt_machine = key[0]
other_machines = [i for i in machines if i != corrupt_machine]
assert len([corrupt_machine] + other_machines) == len(machines)
corrupt_filename = key[1]
block_ops = list(err_map[key])
for (op, block) in block_ops:
possible_err_modes = get_error_modes(op)
for err_type in possible_err_modes:
dir_index = str(corrupt_filename).rfind(data_dirs[corrupt_machine]) + len(data_dirs[corrupt_machine]) + 1
log_dir = 'result_' + (str(corrupt_machine) + '_' + str(corrupt_filename[dir_index :]) + '_' + str(block) + '_' + str(op) + '_' + str(err_type)).replace('/', '_')
log_dir_path = os.path.join(cords_results_base_dir, log_dir)
print str(op) + ' ' + str(corrupt_machine) + ':' + str(corrupt_filename) + ':' + str(block) + ':' + str(err_type)
for mach in machines:
invoke_remote_cmd(machine_ips[mach], "rm -rf " + data_dirs[mach])
invoke_remote_cmd(machine_ips[mach], "cp -R " + data_dir_snapshots[mach] + ' ' + data_dirs[mach])
invoke_remote_cmd(machine_ips[corrupt_machine], "rm -rf " + data_dir_mount_points[corrupt_machine])
invoke_remote_cmd(machine_ips[corrupt_machine], "mkdir " + data_dir_mount_points[corrupt_machine])
fuse_start_command = fuse_command_err%(data_dirs[corrupt_machine], data_dir_mount_points[corrupt_machine], corrupt_filename, block, err_type)
invoke_remote_cmd(machine_ips[corrupt_machine], fuse_start_command)
os.system('sleep 1')
data_dir_curr = []
for mach in machines:
data_dir_curr.append(data_dir_mount_points[mach] if corrupt_machine == mach else data_dirs[mach])
assert len(data_dir_curr) == len(machines)
workload_command_curr = workload_command + " cords "
for ddc in data_dir_curr:
workload_command_curr += ddc + " "
for mach in machines:
workload_command_curr += machine_ips[mach] + " "
workload_command_curr += log_dir_path + " "
os.system("rm -rf " + log_dir_path)
os.system("mkdir -p " + log_dir_path)
os.system("rm -rf /tmp/shoulderr")
os.system("touch /tmp/shoulderr; echo \'fals\' >> /tmp/shoulderr")
(out, err) = invoke_cmd(workload_command_curr)
outfile = os.path.join(log_dir_path, 'workload.out')
os.system("rm -rf " + outfile)
os.system("touch " + outfile)
with open(outfile, 'a') as f:
f.write(out + '\n' + err + '\n')
fuse_unmount = fuse_unmount_command%(data_dir_mount_points[corrupt_machine])
unmount_kill_command = fuse_unmount + '; sleep 1; killall errfs'
invoke_remote_cmd(machine_ips[corrupt_machine], unmount_kill_command)
os.system('mv /tmp/shoulderr ' + log_dir_path)
for mach in machines:
copy_dir_from_remote(machine_ips[mach], data_dirs[mach], log_dir_path + '/')
if replay_check_needed:
checker_command_curr = checker_command + ' ' + log_dir_path
print 'Invoking checker...'
os.system(checker_command_curr)
count += 1
print 'States completed:' + str(count) + '/' + str(total)
if count == 1:
return
start_test = time.time()
cords_check()
print 'cords-check done!'
end_test = time.time()
print 'Testing took ' + str((end_test - start_test)) + ' seconds...'