forked from ureyni/proxysql-admin-tool
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathproxysql_node_monitor
executable file
·486 lines (451 loc) · 29.9 KB
/
proxysql_node_monitor
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
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
#!/bin/bash
# This script will assist to setup Percona XtraDB cluster ProxySQL monitoring script.
#####################################################################################
debug=0
if [ -f /etc/proxysql-admin.cnf ]; then
source /etc/proxysql-admin.cnf
else
echo "`date` Assert! proxysql-admin configuration file :/etc/proxysql-admin.cnf does not exists, Terminating!" >> $ERR_FILE
exit 1
fi
if [[ -z "$PROXYSQL_DATADIR" ]]; then
PROXYSQL_DATADIR='/var/lib/proxysql'
fi
WRITE_HOSTGROUP_ID="${1:-0}"
READ_HOSTGROUP_ID="${2:-0}"
SLAVEREAD_HOSTGROUP_ID=$READ_HOSTGROUP_ID
if [ $SLAVEREAD_HOSTGROUP_ID -eq $WRITE_HOSTGROUP_ID ];then
let SLAVEREAD_HOSTGROUP_ID+=1
fi
ERR_FILE="${3:-/dev/null}"
CHECK_STATUS=0
SLAVE_SECONDS_BEHIND=3600 #How far behind can a slave be before its put into OFFLINE_SOFT state
CLUSTER_TIMEOUT=3 # Maximum time to wait for cluster status
if [ $debug -eq 1 ];then echo "`date` DEBUG MODE: $MODE" >> $ERR_FILE;fi
if [ "$MODE" == "loadbal" ]; then
MODE_COMMENT="READWRITE"
WRITE_WEIGHT="1000"
else
MODE_COMMENT="READ"
WRITE_WEIGHT="1000000"
fi
check_cmd(){
MPID=$1
ERROR_MSG=$2
ERROR_INFO=$3
if [ ${MPID} -ne 0 ]; then
echo "`date` WARNING: $ERROR_MSG." >> $ERR_FILE
if [[ ! -z $ERROR_INFO ]]; then
echo "`date` $ERROR_INFO." >> $ERR_FILE
fi
fi
}
proxysql_exec() {
query=$1
printf "[client]\nuser=${PROXYSQL_USERNAME}\npassword=${PROXYSQL_PASSWORD}\nhost=${PROXYSQL_HOSTNAME}\nport=${PROXYSQL_PORT}\n" | \
mysql --defaults-file=/dev/stdin --protocol=tcp -Bse "${query}" 2>/dev/null
}
#Timeout exists for instances where mysqld may be hung
TIMEOUT=10
mysql_exec() {
query=$1
printf "[client]\nconnect-timeout=${CLUSTER_TIMEOUT}\nuser=${CLUSTER_USERNAME}\npassword=${CLUSTER_PASSWORD}\nhost=${CLUSTER_HOSTNAME}\nport=${CLUSTER_PORT}\n" | \
timeout $TIMEOUT mysql --defaults-file=/dev/stdin --protocol=tcp -Bse "${query}" 2>/dev/null
}
set_slave_status() {
# This function checks the status of slave machines and sets their status field
# The ws_ip and ws_port variables are used and must be set before calling this function
if [ $debug -eq 1 ];then echo "`date` DEBUG START set_slave_status" >> $ERR_FILE;fi
# This function will get and return a status of a slave node, 4=GOOD, 2=BEHIND, 0=OTHER
SLAVE_STATUS=$(printf "[client]\nuser=${CLUSTER_USERNAME}\npassword=${CLUSTER_PASSWORD}\nhost=${ws_ip}\nport=${ws_port}\n" | timeout $TIMEOUT mysql --defaults-file=/dev/stdin --protocol=tcp -Bse 'SHOW SLAVE STATUS\G' 2>${PROXYSQL_DATADIR}/proxysql_admin_error_info)
check_cmd $? "Cannot get status from the slave $ws_ip:$ws_port, Please check cluster login credentials" "`cat ${PROXYSQL_DATADIR}/proxysql_admin_error_info`"
SLAVE_STATUS=$(echo "$SLAVE_STATUS" | sed 's/ //g')
echo "$SLAVE_STATUS" | grep "^Master_Host:" >/dev/null
if [ $? -ne 0 ];then
# No status was found, this is not replicating
if [ $debug -eq 1 ];then echo "`date` DEBUG set_slave_status: No slave status found, setting to OFFLINE_HARD, status was: $ws_status" >> $ERR_FILE;fi
# Only changing the status here as another node might be in the writer hostgroup
proxysql_exec "UPDATE mysql_servers set status = 'OFFLINE_HARD', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${ws_hg_id}:${i} Slave node set to OFFLINE_HARD status to ProxySQL database." >> $ERR_FILE
else
slave_master_host=$(echo "$SLAVE_STATUS" | grep "^Master_Host:" | cut -d: -f2)
slave_io_running=$(echo "$SLAVE_STATUS" | grep "^Slave_IO_Running:" | cut -d: -f2)
slave_sql_running=$(echo "$SLAVE_STATUS" | grep "^Slave_SQL_Running:" | cut -d: -f2)
seconds_behind=$(echo "$SLAVE_STATUS" | grep "^Seconds_Behind_Master:" | cut -d: -f2)
if [ "$seconds_behind" == "NULL" ];then
# When slave_io is not working, the seconds behind value will read 'NULL', convert this to a number higher than the max
let seconds_behind=SLAVE_SECONDS_BEHIND+1
fi
if [ "$slave_io_running" != "Yes" ] && [ "$slave_sql_running" == "Yes" ];then
# Cannot connect to the master
if [ "$ws_status" == "ONLINE" ];then
echo "`date` Slave node (${ws_hg_id}:${i}) This slave cannot connect to it's master: $slave_master_host" >> $ERR_FILE
if [ -z "$CLUSTER_OFFLINE" ];then
# The cluster is up so this slave should go to OFFLINE_SOFT state
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $SLAVEREAD_HOSTGROUP_ID, status = 'OFFLINE_SOFT', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${ws_hg_id}:${i} Slave node set to OFFLINE_SOFT status to ProxySQL database." >> $ERR_FILE
fi
else
if [ -n "$CLUSTER_OFFLINE" ];then
# The slave is not currently online and cannot connect to its master, but we are here because all cluster nodes are down so put the slave ONLINE
if [ $debug -eq 1 ];then echo "`date` DEBUG set_slave_status: Forcing slave $ws_ip:$ws_port ONLINE because cluster is offline" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set status = 'ONLINE', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${SLAVEREAD_HOSTGROUP_ID}:$ws_ip:$ws_port Slave node set to ONLINE status to ProxySQL database." >> $ERR_FILE
else
echo "`date` Slave node (${ws_hg_id}:${i}) current status '$ws_status' in ProxySQL database!" >> $ERR_FILE
fi
fi
elif [ "$slave_sql_running" != "Yes" ];then
# Slave is not replicating
if [ "$ws_status" != "OFFLINE_HARD" ];then
if [ $debug -eq 1 ];then echo "`date` DEBUG set_slave_status: Setting to OFFLINE_HARD, status was: $ws_status" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $SLAVEREAD_HOSTGROUP_ID, status = 'OFFLINE_HARD', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${ws_hg_id}:${i} Slave node set to OFFLINE_HARD status to ProxySQL database." >> $ERR_FILE
else
echo "`date` Slave node (${ws_hg_id}:${i}) current status '$ws_status' in ProxySQL database!" >> $ERR_FILE
fi
elif [ $seconds_behind -gt $SLAVE_SECONDS_BEHIND ];then
# Slave is more than the set number of seconds behind, return status 2
if [ "$ws_status" != "OFFLINE_SOFT" ];then
if [ $debug -eq 1 ];then echo "`date` DEBUG set_slave_status: Setting to OFFLINE_SOFT, status was: $ws_status" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $SLAVEREAD_HOSTGROUP_ID, status = 'OFFLINE_SOFT', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${ws_hg_id}:${i} Slave node set to OFFLINE_SOFT status to ProxySQL database." >> $ERR_FILE
else
echo "`date` Slave node (${ws_hg_id}:${i}) current status '$ws_status' in ProxySQL database!" >> $ERR_FILE
fi
else
if [ "$ws_status" != "ONLINE" ];then
if [ $debug -eq 1 ];then echo "`date` DEBUG set_slave_status: Setting to ONLINE, status was: $ws_status" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set status = 'ONLINE', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${ws_hg_id}:${i} Slave node set to ONLINE status to ProxySQL database." >> $ERR_FILE
else
echo "`date` Slave node (${ws_hg_id}:${i}) current status '$ws_status' in ProxySQL database!" >> $ERR_FILE
fi
fi
fi
if [ $debug -eq 1 ];then echo "`date` DEBUG END set_slave_status" >> $ERR_FILE;fi
}
# Update Percona XtraDB Cluster nodes in ProxySQL database
update_cluster(){
if [ $debug -eq 1 ];then echo "`date` DEBUG START update_cluster" >> $ERR_FILE;fi
current_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers where hostgroup_id in ( $WRITE_HOSTGROUP_ID, $READ_HOSTGROUP_ID, $SLAVEREAD_HOSTGROUP_ID )" | sed 's|\t|:|g' | tr '\n' ' '`)
wsrep_address=(`mysql_exec "SHOW STATUS LIKE 'wsrep_incoming_addresses'" | awk '{print $2}' | sed 's|,| |g'`)
if [ ${#wsrep_address[@]} -eq 0 ]; then
# Cluster might be down, but is there a slave to fall back to?
slave_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers where hostgroup_id in ( $WRITE_HOSTGROUP_ID, $SLAVEREAD_HOSTGROUP_ID ) and comment = 'SLAVEREAD'" | sed 's|\t|:|g' | tr '\n' ' '`)
if [ ${#slave_hosts[@]} -eq 0 ]; then
echo "`date` Alert! wsrep_incoming_addresses is empty. Terminating!" >> $ERR_FILE
exit 1
fi
fi
for i in "${wsrep_address[@]}"; do
if [[ ! " ${current_hosts[@]} " =~ " ${i} " ]]; then
if [ $debug -eq 1 ];then echo "`date` DEBUG Host $i in cluster membership was not found in ProxySQL, adding it" >> $ERR_FILE;fi
ws_ip=$(echo $i | cut -d':' -f1)
ws_port=$(echo $i | cut -d':' -f2)
ws_hg_status=$(echo `proxysql_exec "SELECT hostgroup_id,status from mysql_servers WHERE hostname='$ws_ip' and port=$ws_port"`)
ws_hg_id=$(echo $ws_hg_status | cut -d' ' -f1)
ws_status=$(echo $ws_hg_status | cut -d' ' -f2)
echo "`date` Cluster node (${ws_hg_id}:${i}) does not exists in ProxySQL database!" >> $ERR_FILE
proxysql_exec "INSERT INTO mysql_servers (hostname,hostgroup_id,port,weight,comment) VALUES ('$ws_ip',$READ_HOSTGROUP_ID,$ws_port,1000,'$MODE_COMMENT');"
check_cmd $? "Cannot add Percona XtraDB Cluster node $ws_ip:$ws_port (hostgroup $READ_HOSTGROUP_ID) to ProxySQL database, Please check proxysql credentials"
echo "`date` Added ${ws_hg_id}:${i} node into ProxySQL database." >> $ERR_FILE
CHECK_STATUS=1
fi
done
for i in "${current_hosts[@]}"; do
if [[ ! " ${wsrep_address[@]} " =~ " ${i} " ]]; then
if [ $debug -eq 1 ];then echo "`date` DEBUG Host $i not found in cluster membership" >> $ERR_FILE;fi
# The current host in current_hosts was not found in cluster membership, set it OFFLINE_HARD unless its a slave node
ws_ip=$(echo $i | cut -d':' -f1)
ws_port=$(echo $i | cut -d':' -f2)
ws_hg_status=$(echo `proxysql_exec "SELECT hostgroup_id,status,comment from mysql_servers WHERE hostname='$ws_ip' and port=$ws_port"`)
ws_hg_id=$(echo $ws_hg_status | cut -d' ' -f1)
ws_status=$(echo $ws_hg_status | cut -d' ' -f2)
comment=$(echo $ws_hg_status | cut -d' ' -f3)
if [ "$comment" == "SLAVEREAD" ];then
if [ $debug -eq 1 ];then echo "`date` DEBUG Host $i is a slave, checking its health" >> $ERR_FILE;fi
#This is a slave, check health differently
set_slave_status
else
if [ "$ws_status" == "OFFLINE_SOFT" ]; then
echo "`date` Cluster node ${ws_hg_id}:${i} does not exists in cluster membership! Chaning status from OFFLINE_SOFT to OFFLINE_HARD" >> $ERR_FILE
proxysql_exec "UPDATE mysql_servers set status = 'OFFLINE_HARD', hostgroup_id = $READ_HOSTGROUP_ID, comment='READ', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
CHECK_STATUS=1
fi
node_status=$(echo `proxysql_exec "SELECT status from mysql_servers WHERE hostname='$ws_ip' and port=$ws_port"`)
echo "`date` Cluster node (${ws_hg_id}:${i}) current status '$node_status' in ProxySQL database!" >> $ERR_FILE
if [ "$MODE" == "singlewrite" ]; then
checkwriter_hid=`proxysql_exec "select hostgroup_id from mysql_servers where comment='WRITE' and status='ONLINE'"`
if [[ -z "$checkwriter_hid" ]]; then
current_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='READ' ORDER BY random() LIMIT 1" | sed 's|\t|:|g' | tr '\n' ' '`)
ws_ip=$(echo $current_hosts | cut -d':' -f1)
ws_port=$(echo $current_hosts | cut -d':' -f2)
echo "`date` No writer found, promoting $ws_ip:$ws_port as writer node!" >> $ERR_FILE
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $WRITE_HOSTGROUP_ID, comment='WRITE', weight=1000000 WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
CHECK_STATUS=1
fi
fi
fi
fi
done
for i in "${wsrep_address[@]}"; do
if [[ ! " ${current_hosts[@]} " == " ${i} " ]]; then
if [ $debug -eq 1 ];then echo "`date` DEBUG Host $i was found in cluster membership" >> $ERR_FILE;fi
# current_hosts contains the node in wsrep_addresses
ws_ip=$(echo $i | cut -d':' -f1)
ws_port=$(echo $i | cut -d':' -f2)
ws_hg_status=$(echo `proxysql_exec "SELECT hostgroup_id,status from mysql_servers WHERE hostname='$ws_ip' and port=$ws_port"`)
ws_hg_id=$(echo $ws_hg_status | cut -d' ' -f1)
ws_status=$(echo $ws_hg_status | cut -d' ' -f2)
echo "`date` Cluster node (${ws_hg_id}:${i}) current status '$ws_status' in ProxySQL database!" >> $ERR_FILE
if [ "$ws_status" == "OFFLINE_HARD" ]; then
# The node was OFFLINE_HARD, but its now in the cluster list so lets make it OFFLINE_SOFT
proxysql_exec "UPDATE mysql_servers set status = 'OFFLINE_SOFT', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port;"
check_cmd $? "Cannot update Percona XtraDB Cluster node $ws_ip:$ws_port to ProxySQL database, Please check proxysql credentials"
echo "`date` ${ws_hg_id}:${i} node set to OFFLINE_SOFT status to ProxySQL database." >> $ERR_FILE
CHECK_STATUS=1
fi
fi
done
if [ $debug -eq 1 ];then echo "`date` DEBUG End update_cluster" >> $ERR_FILE;fi
}
mode_change_check(){
if [ $debug -eq 1 ];then echo "`date` DEBUG START mode_change_check" >> $ERR_FILE;fi
if [ -f $HOST_PRIORITY_FILE ];then
# Get the list of hosts from the host_priority file ignoring blanks and any lines that start with '#'
priority_hosts=(`cat $HOST_PRIORITY_FILE | grep ^[^#]`)
fi
# File sample:
# 10.11.12.21:3306
# 10.21.12.21:3306
# 10.31.12.21:3306
#
# Check if the current writer is in an OFFLINE_SOFT state
checkwriter_hid=`proxysql_exec "select hostgroup_id from mysql_servers where comment in ('WRITE', 'READWRITE') and status='OFFLINE_SOFT'"`
if [[ -n "$checkwriter_hid" ]]; then
# Found a writer node that was in 'OFFLINE_SOFT' state, move it to the READ hostgroup unless the MODE is 'loadbal'
if [ "$MODE" != "loadbal" ];then
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check: Found OFFLINE_SOFT writer, changing to READ status and hostgroup $READ_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $READ_HOSTGROUP_ID, comment='READ', weight=1000 WHERE comment='WRITE' and status='OFFLINE_SOFT'"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
CHECK_STATUS=1
fi
# If that temp file exists use it otherwise choose random as is done now
unset current_hosts
if [ -z $priority_hosts ];then
# Order file wasn't found, behave as before
current_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment in ('READ', 'READWRITE') ORDER BY random() LIMIT 1" | sed 's|\t|:|g' | tr '\n' ' '`)
else
# Get the list of all ONLINE reader nodes in ProxySQL
current_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='READ'" | sed 's|\t|:|g' | tr '\n' ' '`)
# Find the highest priority host from the online reader hosts
for i in "${priority_hosts[@]}"; do
if [[ " ${current_hosts[@]} " =~ " ${i} " ]]; then
# This host in priority_hosts was found in the list of current_hosts
current_hosts=${i}
found_host=1
break
fi
done
if [ -z $found_host ];then
# None of the priority hosts were found as active, picking the first on the list from what is available.
current_hosts=(`echo $current_hosts | cut -d' ' -f1`)
unset found_host
fi
fi
# If the $current_hosts variabe is empty here then it's time to put the SLAVEREAD node in if there is one
if [ -z "$current_hosts" ];then
# Verify a slave is not already in the write hostgroup
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check: No cluster members available, check if any slaves are already in the writer hostgroup" >> $ERR_FILE;fi
slave_check=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='SLAVEREAD' and hostgroup_id=$WRITE_HOSTGROUP_ID ORDER BY random() LIMIT 1" | sed 's|\t|:|g' | tr '\n' ' '`)
if [ -z "$slave_check" ];then
# no slaves were currently in the writer group
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check: No slaves currently in the writer group" >> $ERR_FILE;fi
current_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='SLAVEREAD' ORDER BY random() LIMIT 1" | sed 's|\t|:|g' | tr '\n' ' '`)
slave_write="1"
else
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check: A slave is already in the writer hostgroup" >> $ERR_FILE;fi
fi
fi
ws_ip=$(echo $current_hosts | cut -d':' -f1)
ws_port=$(echo $current_hosts | cut -d':' -f2)
# If the cluster is failed and a slave was already in as writer, the current_hosts variable will be empty
if [ "$slave_write" == "1" ] && [ -n "$current_hosts" ];then
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check1: Changing $ws_ip:$ws_port to hostgroup $WRITE_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $WRITE_HOSTGROUP_ID, weight=$WRITE_WEIGHT WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
echo "`date` $ws_ip:$ws_port (slave) is ONLINE, switching to write hostgroup" >> $ERR_FILE
CHECK_STATUS=1
elif [ "$MODE" != "loadbal" ] && [ -n "$current_hosts" ];then
# Only do this if the MODE is not 'loadbal'
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check1: Changing $ws_ip:$ws_port to WRITE status and hostgroup $WRITE_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $WRITE_HOSTGROUP_ID, comment='WRITE', weight=$WRITE_WEIGHT WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
echo "`date` $ws_ip:$ws_port is ONLINE, switching to write hostgroup" >> $ERR_FILE
CHECK_STATUS=1
fi
else
# The current writer was not in OFFLINE_SOFT state
# Now check if the current writer is a slave node and pull it out if other nodes are available
# Should check if a slave is the current writer and pull it out if a cluster node is available
checkslave_hid=`proxysql_exec "select hostgroup_id from mysql_servers where hostgroup_id='$WRITE_HOSTGROUP_ID' AND comment='SLAVEREAD'"`
# Set a variable containing a random available cluster node
available_cluster_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment in ('READ', 'READWRITE') ORDER BY random() LIMIT 1" | sed 's|\t|:|g' | tr '\n' ' '`)
if [[ -n "$checkslave_hid" ]]; then
# The current writer is a slave, check for other ONLINE nodes to put in
if [ -n "$available_cluster_hosts" ];then
# There is a regular cluster node available, pull out the slave
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check1: Changing any SLAVEREAD nodes in hostgroup $WRITE_HOSTGROUP_ID to hostgroup $SLAVEREAD_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $SLAVEREAD_HOSTGROUP_ID, weight=1000 WHERE hostgroup_id='$WRITE_HOSTGROUP_ID' and comment='SLAVEREAD'"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
writer_was_slave=1
fi
fi
if [ -z $priority_hosts ];then
# Order file wasn't found
# Do not change the config of a cluster node if the MODE is 'loadbal'
if [ -n "$available_cluster_hosts" ] && [ -n "$writer_was_slave" ] && [ "$MODE" != "loadbal" ];then
# There is a regular cluster node available, put it back in the writer hostgroup
ws_ip=$(echo $available_cluster_hosts | cut -d':' -f1)
ws_port=$(echo $available_cluster_hosts | cut -d':' -f2)
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check2: Changing $ws_ip:$ws_port to WRITE status and hostgroup $WRITE_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $WRITE_HOSTGROUP_ID, comment='WRITE', weight=$WRITE_WEIGHT WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
echo "`date` $ws_ip:$ws_port is ONLINE, switching to write hostgroup" >> $ERR_FILE
CHECK_STATUS=1
fi
else
# Check here if the highest priority node is the writer
# Get the list of all ONLINE nodes in ProxySQL, can't use available_cluster_hosts here
current_hosts=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' AND comment<>'SLAVEREAD'" | sed 's|\t|:|g' | tr '\n' ' '`)
# Find the highest priority host from the online hosts
for i in "${priority_hosts[@]}"; do
if [[ " ${current_hosts[@]} " =~ " ${i} " ]]; then
# This host in priority_hosts was found in the list of current_hosts
current_hosts=${i}
found_host=1
break
fi
done
# Only initiate changing hosts if a more priority host was found
if [ -n $found_host ];then
# Check to see if the host in 'current_host' is the writer
current_writer=(`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='WRITE'" | sed 's|\t|:|g' | tr '\n' ' '`)
if [ "$current_hosts" != "$current_writer" ];then
# Switch the writer around
if [ -n "$current_writer" ];then
# Move the current writer host to reader hostgroup if there is one
ws_ip=$(echo $current_writer | cut -d':' -f1)
ws_port=$(echo $current_writer | cut -d':' -f2)
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check3: Changing $ws_ip:$ws_port to READ status and hostgroup $READ_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $READ_HOSTGROUP_ID, comment='READ', weight=1000 WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster reader node in ProxySQL database, Please check proxysql credentials"
echo "`date` $ws_ip:$ws_port is ONLINE but a higher priority node is available, switching to read hostgroup" >> $ERR_FILE
fi
# Move the priority host to the writer hostgroup
ws_ip=$(echo $current_hosts | cut -d':' -f1)
ws_port=$(echo $current_hosts | cut -d':' -f2)
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check3: Changing $ws_ip:$ws_port to WRITE status and hostgroup $WRITE_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $WRITE_HOSTGROUP_ID, comment='WRITE', weight=$WRITE_WEIGHT WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
echo "`date` $ws_ip:$ws_port is ONLINE and highest priority, switching to write hostgroup" >> $ERR_FILE
CHECK_STATUS=1
fi
else
if [ -n "$available_cluster_hosts" ] && [ -n "$writer_was_slave" ];then
# There is a regular cluster node available, pull out the slave and put the cluster node back in the writer hostgroup
ws_ip=$(echo $available_cluster_hosts | cut -d':' -f1)
ws_port=$(echo $available_cluster_hosts | cut -d':' -f2)
if [ $debug -eq 1 ];then echo "`date` DEBUG mode_change_check4: Changing $ws_ip:$ws_port to hostgroup $WRITE_HOSTGROUP_ID" >> $ERR_FILE;fi
proxysql_exec "UPDATE mysql_servers set hostgroup_id = $WRITE_HOSTGROUP_ID, weight=$WRITE_WEIGHT WHERE hostname='$ws_ip' and port=$ws_port"
check_cmd $? "Cannot update Percona XtraDB Cluster writer node in ProxySQL database, Please check proxysql credentials"
echo "`date` $ws_ip:$ws_port is ONLINE, switching to write hostgroup" >> $ERR_FILE
CHECK_STATUS=1
fi
fi
fi
fi
if [ $debug -eq 1 ];then echo "`date` DEBUG END mode_change_check" >> $ERR_FILE;fi
}
# Monitoring user needs 'REPLICATION CLIENT' privilege
echo "`date` ###### Percona XtraDB Cluster status ######" >> $ERR_FILE
CLUSTER_USERNAME=$(proxysql_exec "SELECT variable_value FROM global_variables WHERE variable_name='mysql-monitor_username'")
check_cmd $? "Could not retrieve cluster login info from ProxySQL. Please check cluster login credentials"
CLUSTER_PASSWORD=$(proxysql_exec "SELECT variable_value FROM global_variables WHERE variable_name='mysql-monitor_password'")
check_cmd $? "Could not retrieve cluster login info from ProxySQL. Please check cluster login credentials"
CLUSTER_HOST_INFO=`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment<>'SLAVEREAD' and hostgroup_id in ($WRITE_HOSTGROUP_ID, $READ_HOSTGROUP_ID) limit 1"`
check_cmd $? "Could not retrieve cluster node info from ProxySQL. Please check cluster login credentials"
CLUSTER_HOSTS=($(proxysql_exec "SELECT hostname || '-' || port FROM mysql_servers WHERE status='ONLINE' and comment<>'SLAVEREAD' and hostgroup_id in ($WRITE_HOSTGROUP_ID, $READ_HOSTGROUP_ID)"))
CLUSTER_TIMEOUT=($(proxysql_exec "SELECT MAX(interval_ms / 1000 - 1, 1) FROM scheduler"))
for i in "${CLUSTER_HOSTS[@]}"; do
CLUSTER_HOSTNAME=$(echo $i | cut -d'-' -f1)
CLUSTER_PORT=$(echo $i | cut -d'-' -f2)
CHECK_SERVER_STATUS=$(mysql_exec "SELECT @@port")
if [[ -n $CHECK_SERVER_STATUS ]]; then
CLUSTER_HOST_INFO="${CHECK_SERVER_STATUS}"
break
else
CLUSTER_HOST_INFO=""
fi
done
if [[ -z $CLUSTER_HOST_INFO ]]; then
if [ $debug -eq 1 ];then echo "`date` DEBUG Can't get cluster info, checking if a slave is available" >> $ERR_FILE;fi
# Set CLUSTER_OFFLINE variable, used my the set_slave_status function and the bottom of this script
CLUSTER_OFFLINE=1
# No Cluster nodes are available, but is a slave available?
SLAVE_CHECK=`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE comment='SLAVEREAD' limit 1"`
if [[ ! -z $SLAVE_CHECK ]]; then
SLAVE_HOST_INFO=`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='SLAVEREAD' and hostgroup_id in ($WRITE_HOSTGROUP_ID, $SLAVEREAD_HOSTGROUP_ID) limit 1"`
check_cmd $? "Could not retrieve cluster login info from ProxySQL. Please check cluster login credentials"
if [[ -z $SLAVE_HOST_INFO ]]; then
if [ $debug -eq 1 ];then echo "`date` DEBUG No online slaves were found, will recheck" >> $ERR_FILE;fi
# Check for a slave in a status other than 'ONLINE'
# This is an emergency measure, just put a random slave online
# Would be nice to try to find the most up to date slave if there is more than one, but that would require
# a query to all slaves to check their positions, probably not worth the overhead - something to think about
slave_host=(`proxysql_exec "SELECT hostname,port FROM mysql_servers where comment='SLAVEREAD' and hostgroup_id in ($WRITE_HOSTGROUP_ID, $SLAVEREAD_HOSTGROUP_ID) ORDER BY random() LIMIT 1" | sed 's|\t|:|g' | tr '\n' ' '`)
if [ $debug -eq 1 ];then echo "`date` DEBUG Trying to bring slave: $slave_host ONLINE due to cluster being down" >> $ERR_FILE;fi
ws_ip=$(echo $slave_host | cut -d':' -f1)
ws_port=$(echo $slave_host | cut -d':' -f2)
set_slave_status
else
if [ $debug -eq 1 ];then echo "`date` DEBUG online slaves were found" >> $ERR_FILE;fi
# Run function here to move the slave into the write hostgroup
mode_change_check
fi
fi
SLAVE_HOST_INFO=`proxysql_exec "SELECT hostname,port FROM mysql_servers WHERE status='ONLINE' and comment='SLAVEREAD' and hostgroup_id='$WRITE_HOSTGROUP_ID' limit 1"`
if [[ -z $SLAVE_HOST_INFO ]]; then
offline_hosts=(`proxysql_exec "SELECT hostgroup_id,hostname,port,status FROM mysql_servers where hostgroup_id in ( $WRITE_HOSTGROUP_ID, $READ_HOSTGROUP_ID, $SLAVEREAD_HOSTGROUP_ID )" | sed 's|\t|:|g' | tr '\n' ' '`)
for i in "${offline_hosts[@]}"; do
offline_host=(`echo ${i} | awk -F: '{print $1 ":" $2 ":" $3}'`)
offline_host_status=(`echo ${i} | cut -d':' -f4`)
echo "`date` Cluster node ($offline_host) current status '$offline_host_status' in ProxySQL database!" >> $ERR_FILE
done
else
echo "`date` Percona XtraDB Cluster nodes are offline, a slave node is in the writer hostgroup, please check status" >> $ERR_FILE
fi
else
update_cluster
mode_change_check
fi
if [ $CHECK_STATUS -eq 0 ]; then
if [ -z "$CLUSTER_OFFLINE" ];then
echo "`date` Percona XtraDB Cluster membership looks good" >> $ERR_FILE
else
echo "`date` Percona XtraDB Cluster is offline!" >> $ERR_FILE
fi
else
echo "`date` ###### Loading mysql_servers config into runtime ######" >> $ERR_FILE
proxysql_exec "LOAD MYSQL SERVERS TO RUNTIME"
fi
exit 0