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
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
|
# See the file LICENSE for redistribution information.
#
# Copyright (c) 2004,2007 Oracle. All rights reserved.
#
# $Id: rep069.tcl,v 12.5 2007/06/15 14:39:51 carol Exp $
#
# TEST rep069
# TEST Test of internal initialization and elections.
# TEST
# TEST If a client is in a recovery mode of any kind, it
# TEST participates in elections at priority 0 so it can
# TEST never be elected master.
#
proc rep069 { method { niter 200 } { tnum "069" } args } {
source ./include.tcl
if { $is_windows9x_test == 1 } {
puts "Skipping replication test on Win 9x platform."
return
}
if { $checking_valid_methods } {
set test_methods { btree }
return $test_methods
}
if { [is_btree $method] == 0 } {
puts "Rep$tnum: Skipping for method $method."
return
}
set args [convert_args $method $args]
set saved_args $args
set logsets [create_logsets 2]
foreach l $logsets {
set args $saved_args
puts "Rep$tnum ($method $args):\
Test internal initialization and elections."
puts "Rep$tnum: Master logs are [lindex $l 0]"
puts "Rep$tnum: Client logs are [lindex $l 1]"
rep069_sub $method $niter $tnum $l $args
}
}
proc rep069_sub { method niter tnum logset largs } {
global testdir
global util_path
global rep_verbose
global timeout_ok
env_cleanup $testdir
set qdir $testdir/MSGQUEUEDIR
replsetup $qdir
set verbargs ""
if { $rep_verbose == 1 } {
set verbargs " -verbose { rep on } "
}
set masterdir $testdir/MASTERDIR
file mkdir $masterdir
set nclients 2
for { set i 0 } { $i < $nclients } { incr i } {
set clientdir($i) $testdir/CLIENTDIR.$i
file mkdir $clientdir($i)
}
# Log size is small so we quickly create more than one, and
# can easily force internal initialization.
set pagesize 4096
append largs " -pagesize $pagesize "
set log_max [expr $pagesize * 8]
set m_logtype [lindex $logset 0]
set c_logtype [lindex $logset 1]
# In-memory logs cannot be used with -txn nosync.
set m_logargs [adjust_logargs $m_logtype]
set c_logargs [adjust_logargs $c_logtype]
set m_txnargs [adjust_txnargs $m_logtype]
set c_txnargs [adjust_txnargs $c_logtype]
# Open a master.
set envlist {}
repladd 1
set ma_envcmd "berkdb_env_noerr -create $m_txnargs \
$m_logargs -log_max $log_max -event rep_event $verbargs \
-home $masterdir -rep_transport \[list 1 replsend\]"
set masterenv [eval $ma_envcmd -recover -rep_master]
lappend envlist "$masterenv 1"
# Open clients.
for { set i 0 } { $i < $nclients } { incr i } {
set envid [expr $i + 2]
repladd $envid
set envcmd($i) "berkdb_env_noerr -create \
$c_txnargs $c_logargs -log_max $log_max \
-home $clientdir($i) -event rep_event $verbargs \
-rep_transport \[list $envid replsend\]"
set clientenv($i) [eval $envcmd($i) -recover -rep_client]
lappend envlist "$clientenv($i) $envid"
}
# Bring the clients online by processing the startup messages.
process_msgs $envlist
# Clobber replication's 30-second anti-archive timer, which will have
# been started by client sync-up internal init, so that we can do a
# db_archive in a moment.
#
$masterenv test force noarchive_timeout
# Run rep_test in the master and update clients.
puts "\tRep$tnum.a: Running rep_test in replicated env."
set start 0
eval rep_test \
$method $masterenv NULL $niter $start $start 0 0 $largs
incr start $niter
process_msgs $envlist 0 NONE err
error_check_good process_msgs $err 0
# Find out what exists on the client. We need to loop until
# the first master log file > last client log file. The two
# clients should be the same, so just inspect one.
puts "\tRep$tnum.b: Close clients."
if { $c_logtype != "in-memory" } {
set res [eval exec $util_path/db_archive -l -h $clientdir(0)]
set res [eval exec $util_path/db_archive -l -h $clientdir(1)]
}
set last_client_log [get_logfile $clientenv(1) last]
for { set i 0 } { $i < $nclients } { incr i } {
error_check_good client_close [$clientenv($i) close] 0
}
set envlist [lreplace $envlist 1 2]
# Run the master forward.
set stop 0
while { $stop == 0 } {
puts "\tRep$tnum.c: Running rep_test in replicated env."
eval rep_test \
$method $masterenv NULL $niter $start $start 0 0 $largs
incr start $niter
puts "\tRep$tnum.d: Run db_archive on master."
if { $m_logtype != "in-memory"} {
set res [eval \
exec $util_path/db_archive -d -h $masterdir]
}
set first_master_log [get_logfile $masterenv first]
if { $first_master_log > $last_client_log } {
set stop 1
}
}
for { set i 0 } { $i < $nclients } { incr i } {
set envid [expr $i + 2]
replclear $envid
}
# Reopen clients.
puts "\tRep$tnum.e: Reopen clients."
for { set i 0 } { $i < $nclients } { incr i } {
env_cleanup $clientdir($i)
set clientenv($i) [eval $envcmd($i) -recover -rep_client]
set envid [expr $i + 2]
lappend envlist "$clientenv($i) $envid"
}
# Run proc_msgs_once until both clients are in internal
# initialization.
#
# We figure out whether each client is in initialization
# by searching for any of the flags REP_F_RECOVER_UPDATE,
# REP_F_RECOVER_PAGE, and REP_F_RECOVER_LOG. As soon as
# a client produces one of these, it's marked as being
# in initialization, and stays that way even if it proceeds
# further, but we don't exit the loop until all clients
# have gotten into initialization.
#
puts "\tRep$tnum.f:\
Run proc_msgs_once until all clients enter internal init."
set in_init 0
for { set i 0 } { $i < $nclients } { incr i } {
set initializing($i) 0
}
while { $in_init != 1 } {
set nproced [proc_msgs_once $envlist NONE err]
for { set i 0 } { $i < $nclients } { incr i } {
set stat($i) \
[exec $util_path/db_stat -r -R A -h $clientdir(1)]
if {[is_substr $stat($i) "REP_F_RECOVER_UPDATE"] } {
set initializing($i) 1
}
if {[is_substr $stat($i) "REP_F_RECOVER_PAGE"] } {
set initializing($i) 1
}
if {[is_substr $stat($i) "REP_F_RECOVER_LOG"] } {
set initializing($i) 1
}
}
set in_init 1
for { set i 0 } { $i < $nclients } { incr i } {
if { $initializing($i) == 0 } {
set in_init 0
}
}
}
# Call an election. It should fail, because both clients
# are in internal initialization and therefore not electable.
# Indicate failure with winner = -2.
# First, close the master.
error_check_good masterenv_close [$masterenv close] 0
set envlist [lreplace $envlist 0 0]
puts "\tRep$tnum.g: Run election; no one will get elected."
set m "Rep$tnum.g"
set nsites $nclients
set nvotes $nclients
set winner -2
set elector 0
for { set i 0 } { $i < $nclients } { incr i } {
set err_cmd($i) "none"
set crash($i) 0
set pri($i) 10
}
# This election will time out instead of succeeding.
set timeout_ok 1
run_election envcmd envlist err_cmd pri crash \
$qdir $m $elector $nsites $nvotes $nclients $winner \
0 "test.db" 0 $timeout_ok
# Verify that each client saw the message that no
# electable site was found.
puts "\tRep$tnum.h: Check for right error message."
for { set i 0 } { $i < $nclients } { incr i } {
set none_electable 0
set id [expr $i + 1]
set fd [open $testdir/ELECTION_ERRFILE.$id r]
while { [gets $fd str] != -1 } {
if { [is_substr $str "No electable site found"] == 1 } {
set none_electable 1
break
}
}
close $fd
error_check_good none_electable $none_electable 1
}
# Clean up for the next pass.
for { set i 0 } { $i < $nclients } { incr i } {
$clientenv($i) close
}
replclose $testdir/MSGQUEUEDIR
}
|