Error 1146: Table 'replication_manager_schema.pseudo_gtid_v' doesn't exist

[nigel889]

used replication-manager-pro-2.0.0_20_g97b7a-1.x86_64.rpm installed
percona5.6.32 gtid mode = off
1master --> 2 slave
master: 10.1.1.173
slave1: 10.1.1.171
slave2: 10.1.1.172

when switchover, mrm show logs:
Could not get pseudoGTID on slave 10.1.1.172:3306, Error 1146: Table 'replication_manager_schema.pseudo_gtid_v' doesn't exist
10.1.1.172:3306 Slave_SQL_Running: no ,replication stop.
how to fix it.

[svaroqui]

Hello, Are you using replication-filters like do-db or ignore-db ? /svar

Le 17 mai 2018 à 13:37, nigel889 ***@***.***> a écrit : used replication-manager-pro-2.0.0_20_g97b7a-1.x86_64.rpm installed percona5.6.32 gtid mode = off 1master --> 2 slave master: 10.1.1.173 slave1: 10.1.1.171 slave2: 10.1.1.172 when switchover, mrm show logs: Could not get pseudoGTID on slave 10.1.1.172:3306, Error 1146: Table 'replication_manager_schema.pseudo_gtid_v' doesn't exist 10.1.1.172:3306 Slave_SQL_Running: no ,replication stop. how to fix it. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#235>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AC1RIEw68_m9vyE5Vxl_WI6wX6ZsLBYIks5tzWDdgaJpZM4UC5nl>.

Stéphane Varoqui, VP of Products Phone: +33 695-926-401, skype: svaroqui https://signal18.io/ <https://signal18.io/>

[nigel889]

my.cnf setting replicate-wild-ignore-table :

replicate-wild-ignore-table = mysql.%
replicate-wild-ignore-table = information_schema.%
replicate-wild-ignore-table = performance_schema.%
replicate-wild-ignore-table = temp.%

[nigel889]

my.cnf remove replicate-wild-ignore-table:
#replicate-wild-ignore-table = mysql.%
#replicate-wild-ignore-table = information_schema.%
#replicate-wild-ignore-table = performance_schema.%
#replicate-wild-ignore-table = temp.%
and restart mysql service , when mrm swichover, logs report:

2018/05/17 22:55:39 [pub] ERROR - Could not get pseudoGTID on slave 10.1.1.173:3306, Error 1146: Table 'replication_manager_schema.pseudo_gtid_v' doesn't exist
2018/05/17 22:55:39 [pub] INFO - Found pseudoGTID
2018/05/17 22:55:39 [pub] INFO - Found Coordinates on slave mysql-bin.000007, 4
2018/05/17 22:55:39 [pub] ERROR - Could not find number of events after pseudoGTID in slave 10.1.1.173:3306, Error 1220: Error when executing command SHOW BINLOG EVENTS: Could not find target log
2018/05/17 23:10:43 [pub] ERROR - Could not get pseudoGTID on slave 10.1.1.172:3306, Error 1146: Table 'replication_manager_schema.pseudo_gtid_v' doesn't exist
2018/05/17 23:10:43 [pub] INFO - Found pseudoGTID
2018/05/17 23:10:43 [pub] INFO - Found Coordinates on slave mysql-bin.000001, 4
2018/05/17 23:10:43 [pub] ERROR - Could not find number of events after pseudoGTID in slave 10.1.1.172:3306, Error 1220: Error when executing command SHOW BINLOG EVENTS: Could not find target log
2018/05/17 23:10:43 [pub] INFO - Found 0 events to skip after coordinates on slave mysql-bin.000001,4
2018/05/17 23:10:43 [pub] INFO - Found coordinate on master mysql-bin.000001 ,4
2018/05/17 23:10:43 [pub] INFO - Found skip coordinate on master mysql-bin.000001, 4

swichover and failover not work.

[svaroqui]

Re, Sorry was outside for a while ! Think you should have bin logs enable and log_slave_updates on all cluster nodes !

Le 17 mai 2018 à 17:01, nigel889 ***@***.***> a écrit : my.cnf remove replicate-wild-ignore-table: #replicate-wild-ignore-table = mysql.% #replicate-wild-ignore-table = information_schema.% #replicate-wild-ignore-table = performance_schema.% #replicate-wild-ignore-table = temp.% and restart mysql service , when mrm swichover, logs report: 2018/05/17 22:55:39 [pub] ERROR - Could not get pseudoGTID on slave 10.1.1.173:3306, Error 1146: Table 'replication_manager_schema.pseudo_gtid_v' doesn't exist 2018/05/17 22:55:39 [pub] INFO - Found pseudoGTID 2018/05/17 22:55:39 [pub] INFO - Found Coordinates on slave mysql-bin.000007, 4 2018/05/17 22:55:39 [pub] ERROR - Could not find number of events after pseudoGTID in slave 10.1.1.173:3306, Error 1220: Error when executing command SHOW BINLOG EVENTS: Could not find target log — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#235 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AC1RIGYcR7LzsDYNSuoF-wndB1O4QF9lks5tzZDEgaJpZM4UC5nl>.

Stéphane Varoqui, VP of Products Phone: +33 695-926-401, skype: svaroqui https://signal18.io/ <https://signal18.io/>

[nigel889]

Think you should have bin logs enable and log_slave_updates on all cluster nodes !
yes, all cluster nodes :
log_slave_updates = 1 and log_bin enable

[nigel889]

I enable general_log at master , but don't find create pseudoGTID sql in general log, how the pseudoGTID have been created?

i test this case, if disable pseudoGTID , switchover ok, but when enable pseudoGTID, Could not find number of events after pseudoGTID, change master to not correct, failed.

failover-falsepositive-heartbeat = true , no Inject PseudoGTID , don't take effect ?

[svaroqui]

You are making good progress, it goes via a proxy , as you don't have proxy but vip please use extproxy-address

[nigel889]

thanks svaroqui: i want to continue this case. What should I do? can you help me?

[svaroqui]

Are you using a VIP or a Supported Proxy ?
if VIP you should configure

extproxy = true
extproxy-address="VIP_ADDR:3306"

[nigel889]

if enable pseudoGTID , must use proxy or vip ?

[svaroqui]

You can in theory use both i tested with proxy at that time!

[svaroqui]

db, err := cluster.GetClusterThisProxyConn(pr)
		if err != nil {
			cluster.sme.AddState("ERR00050", state.State{ErrType: "ERROR", ErrDesc: fmt.Sprintf(clusterError["ERR00050"], err), ErrFrom: "TOPO"})
		} else {
			if pr.Type == proxyMyProxy {
				definer = "DEFINER = root@localhost"
			} else {
				definer = ""
			}
			_, err := db.Exec("CREATE OR REPLACE " + definer + " VIEW replication_manager_schema.pseudo_gtid_v as select '" + misc.GetUUID() + "' from dual")

			if err != nil {
				cluster.sme.AddState("ERR00050", state.State{ErrType: "ERROR", ErrDesc: fmt.Sprintf(clusterError["ERR00050"], err), ErrFrom: "TOPO"})
				db.Exec("CREATE DATABASE IF NOT EXISTS replication_manager_schema")

			}
			db.Close()
		}

[svaroqui]

there is may be create or replace i hope your DB support such syntax ?

[nigel889]

whether use master's ip:port of cluster instead of the vip:port or proxy 's ip:port?

[svaroqui]

GetClusterThisProxyConn is taken on each proxy
for _, pr := range cluster.Proxies {

And VIP is one of them
extproxy = true
extproxy-address="VIP_ADDR:3306"

Proxy can be maxscale ,shardproxy (mariadb+spider), proxysql , haproxy or myproxy and internal vitess proxy (limited in syntax)