MySQLTalk.org

Andrew Garrett · Guest

Thanks Martin, and everyone else who replied off-list to let me know I
was in the right place.

So, here's my situation

I'm trying to set up a fairly bog-standard single master -> single
slave replication (for now, at least).

After doing what the docs tell me to
(http://dev.mysql.com/doc/refman/5.0/en/replication-howto.html), these
are the symptoms:

On the master: mysql> show master status\G;

*************************** 1. row ***************************

File: mysql-bin.000110

Position: 14817654

Binlog_Do_DB: jaiku_prod,jaikupresence_prod,mqueue_prod

Binlog_Ignore_DB: jaiku_dev,jaikupresence_dev,mqueue_dev,mysql 1
row in set (0.00 sec)

On the slave: mysql> show slave status\G;

*************************** 1. row ***************************

Slave_IO_State: Waiting for master to send event

Master_Host: 83.145.232.225 Master_User: slave_user Master_Port: 3306

Connect_Retry: 60

Master_Log_File: mysql-bin.000110

Read_Master_Log_Pos: 14814780

Relay_Log_File: kaksi-relay-bin.000004

Relay_Log_Pos: 14814917

Relay_Master_Log_File: mysql-bin.000110

Slave_IO_Running: Yes

Slave_SQL_Running: Yes

Replicate_Do_DB: jaiku_prod, jaikupresence_prod, mqueue_prod

Replicate_Ignore_DB:

Replicate_Do_Table:

Replicate_Ignore_Table:

Replicate_Wild_Do_Table:

Replicate_Wild_Ignore_Table:

Last_Errno: 0 Last_Error:

Skip_Counter: 0

Exec_Master_Log_Pos: 14814780

Relay_Log_Space: 14814917

Until_Condition: None

Until_Log_File:

Until_Log_Pos: 0

Master_SSL_Allowed: No Master_SSL_CA_File:

Master_SSL_CA_Path:

Master_SSL_Cert:

Master_SSL_Cipher:

Master_SSL_Key:

Seconds_Behind_Master: 0

1 row in set (0.00 sec)

As far as I can tell, these say, basically, 'every thing is normal,
the slave is up to date.' (looking in particular at which binlog it's
using, and the various 'position' (POS) markers.

However, on the master:

mysql> update user set mobile='+64212266901' where nick='malach';

A real simple change (that's my phone number, incremented by 1.).

On the slave:

Pause long enough to retype the command, bearing in mind that the
slave is telling us it's 0 seconds behind.

mysql> select mobile from user where nick='malach';

+--------------+
| mobile |
+--------------+
| +64212266900 |
+--------------+
1 row in set (0.00 sec)

Wait a while... a few minute... go make a coffee, whatever. Come back,
re-run the select - still the same. The update doesn't make it to the
slave.

No errors in any log I can find. All config options pretty much stock
from the debian packagesm, with the exception of the replication
changes.

$ mysql --version
mysql Ver 14.12 Distrib 5.0.32, for pc-linux-gnu (i486) using readline 5.2

On both hosts

For now, I'm stumped.

This is also my first replication setup in a real world scenario, so I
may have done something painfully dumb.

Any and all assistance will be much appreciated.

Thanks,

Andrew

On 3/25/07, Martin MC Brown <mc (AT) mysql (DOT) com> wrote:

Augusto Bott · Guest

hm... i know these may be 'dumb questions', but i must ask...

are you using transaction-safe-tables on this setup? did you commit
your transactions on your master? (nothings gets written to the binlog
unless it's committed, right?) did you start a new transaction to
retrieve the updated row on the slave *after* the transaction has been
run locally?
can you read the binlog in the master to check if the transaction has
been written? (you can use the command 'mysqlbinlog filename.bin' for
that). i've noticed you are using this on the master:

Binlog_Do_DB: jaiku_prod,jaikupresence_prod,mqueue_prod
Binlog_Ignore_DB: jaiku_dev,jaikupresence_dev,mqueue_dev,mysql

according to http://dev.mysql.com/doc/refman/5.0/en/binary-log.html, i
must ask: did you issue a USE command before updating the row?

anyways... good luck for you :-)

--
Augusto Bott

On 3/25/07, Andrew Garrett <malach (AT) gmail (DOT) com> wrote:

Ed Pauley II · Guest

Andrew Garrett wrote: