>
>
> If you are finding replication stops after a reboot then I think you have a
> serious problem, but don't know what that might be. Post some more info and
> someone on the list can probably help,
>
> just a 'skip-slave-start' in my.cnf does the trick.
2> Is there a way to get a slave to auto
>> But EXPLAIN is only a prediction. If you look at the changes in the
>> Handler status variables, you'll see the second one reads fewer rows.
>>
>>
>
> Ok, I think I get it. I first changed both of my queries to add
> "sql_no_cache" because without that, the Handler_read_rnd_next variable was
>
BTW there is a list specially dedicated to replication configurations:
http://lists.mysql.com/replication
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/mysql?unsub=arch...@jab.org
:S
Hi Bryan
Actually changing the slave setup after a reboot is a very bad
idea, you will have inconsistent data on your slave because every
transaction made from when you rebooted the server to when you reset
the position will not be copied from the master. Thats the point of
the posi
The information is there and the replica can connect on a reboot but
only if that information has not changed correct?
e.g. I reboot the slave, and on the master just run "reset master;"
the bin.01 could change to bin.02 and the pos change from 98
to 15443 or some such. If I do that do I
When you start up replication the data should be recorded in the
master.info in your data dir.
[r...@someserver ~]# cat /var/lib/mysqldata/master.info
14
MASTERHOST-bin.01
MASTERPOS
192.168.0.2
REPLICATION_USER
REPLICATION_PASSWORD
3306
60
0
I am not familiar with any setups where the master.
Do I jsut need to monitor better and manually add it should the log
and pos change?
-Bryan
On Wed, Mar 11, 2009 at 7:05 PM, Baron Schwartz wrote:
> That's deprecated too :-)
>
> On Wed, Mar 11, 2009 at 9:17 PM, Cui Shijun wrote:
>> hi,
>> #2. try adding the information of master into my.cnf t
I have a view which is a 3 table join on a compound index.
I have two indexes: Index1: Product_Code, Store_Id, Date_Sold and
Index2:
Date_Sold,Store_Id,Product_Code
If I execute a select like:
select * from MyView where product_code="123";
it returns
Baron Schwartz wrote:
Hi!
On Wed, Mar 11, 2009 at 8:20 PM, David M. Karr
wrote:
Using 5.0.67-0ubuntu6 on Ubuntu 8.10.
I'm going through the "High Performance MySQL" book. I was reading section
4.4.1.8, titled "MIN() and MAX()". The point of this is that MySQL doesn't
optimize MIN()/MAX()
The nice thing about InnnoDB is that it won't have to access the data
portion of the file if it doesn't have to. So if all the information
you are retrieving is contained in an index, it only accesses the
index to get the information it needs. The data portion is never
access, and thus never locked
Brent,
After a delay while I was busy killing alligators, I did as you suggested
(added a composite index of date and organization_serial on
journal_entry_master... in the spirit of your suggestion, anyway.) The
results were interesting:
1. In my test environment, I could not force a locke
Hi,
My understanding is that the memory utilization of mysql can be calculated
roughly using the formula like:
(All global memory related server variables + max_connections * session
memory related server variables)
As I noticed that most global variables like key_buffer_size,
innodb_buffer_szie w
On Mon, Mar 9, 2009 at 9:10 AM, Stefan Hinz wrote:
> MySQL and ZFS
> http://forge.mysql.com/wiki/MySQL_and_ZFS
>
> This Thursday (March 12th, 14:00 UTC), Martin MC Brown will give a MySQL
> University session on MySQL and ZFS. MC works on the MySQL Documentation
> Team and has been involved with q
Dear Sir,
I am using Mysql 5 in windows system.
I have formatted my system and copy the data directory before formatting the
system.
After formatting the system I have installed Mysql5 and resorted the data
directory with the older one.
When I browsing the database then innodb tabl
14 matches
Mail list logo
