tag:blogger.com,1999:blog-16959946.post3654504716439071711..comments2023-12-09T16:44:47.897+01:00Comments on The Data Charmer: Getting started with MySQL 5.5Giuseppe Maxiahttp://www.blogger.com/profile/15801583338057324813noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-16959946.post-22666269738634593942010-09-23T23:01:09.842+02:002010-09-23T23:01:09.842+02:00It would help other people finding this blog (and ...It would help other people finding this blog (and other introductions to semi-sync replication) if you made it clear that you have to stop and re-start the slave IO_THREAD after setting rpl_semi_sync_slave_enabled=1.<br /><br />You might also want to feed back to the documentation team that the enabled variables also default to 0, not 1 as it says at http://dev.mysql.com/doc/refman/5.5/en/replication-semisync-installation.html.<br /><br />With that, I'm finding semi-sync replication works exactly as expected on Windows Server 2008 with MySQL 5.5.6rc - great work by the MySQL team and thanks also to the Google guys who initiated this a few years ago.camtonynoreply@blogger.comtag:blogger.com,1999:blog-16959946.post-73937384808255381142010-07-10T22:08:09.719+02:002010-07-10T22:08:09.719+02:00Its already the mid of 2010.. when is the GA relea...Its already the mid of 2010.. when is the GA release of MySQL 5.5 expected??Varunhttps://www.blogger.com/profile/11648034621030086321noreply@blogger.comtag:blogger.com,1999:blog-16959946.post-76446798201642190732010-01-12T17:11:54.919+01:002010-01-12T17:11:54.919+01:00what do you call normal replication ? You said nor...what do you call normal replication ? You said normal replication would not work if we stop thread SQL_THREAD, but if we use standard asynchronous replication the statement would be in the relay log too.Cyril Scetbonhttps://www.blogger.com/profile/05787792882719474833noreply@blogger.comtag:blogger.com,1999:blog-16959946.post-51058467657859949472009-12-18T08:48:38.826+01:002009-12-18T08:48:38.826+01:00Jon,
you are right about LIST. I was tempted to me...Jon,<br />you are right about LIST. I was tempted to mention it, and then I decided to link to the manual instead. If I say everything in my blog, I would deprive readers of enjoying all the entertaining pages you have written on the manual about partitions. So, don't complain. I am working for you!<br /><br />Cheers<br /><br />GiuseppeGiuseppe Maxiahttps://www.blogger.com/profile/15801583338057324813noreply@blogger.comtag:blogger.com,1999:blog-16959946.post-80472270087553253272009-12-18T08:19:38.622+01:002009-12-18T08:19:38.622+01:00You neglected to mention that the COLUMNS extensio...You neglected to mention that the COLUMNS extension can also be used with PARTITION BY LIST, not just PARTITION BY RANGE. <br /><br />Also, why does your example show only *2* partitions? That's a bit sparse, isn't it? And might some folks take that the wrong way...?Jon Stephensnoreply@blogger.comtag:blogger.com,1999:blog-16959946.post-79861308379525404682009-12-15T16:13:32.076+01:002009-12-15T16:13:32.076+01:00The MySQL replication team did a great job on semi...The MySQL replication team did a great job on semi-sync. At companies that maintain forks, we must also optimize for patch/fork maintenance and don't always build things the <b>right</b> way. The MySQL team did this the right way and rewrote it using the Google patch as a guide. The result is higher code quality.Mark Callaghanhttps://www.blogger.com/profile/09590445221922043181noreply@blogger.com