tag:blogger.com,1999:blog-6551499576288463009.post4001157874945337614..comments2017-01-18T05:49:54.945-05:00Comments on MySQL and LAMP related posts: Delayed row-based replication with large tables lacking a primary keyAnonymoushttp://www.blogger.com/profile/10372091392084104297noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-6551499576288463009.post-34156905042157788542013-05-15T20:15:27.193-04:002013-05-15T20:15:27.193-04:00Good catch. I have corrected the sentence to prope...Good catch. I have corrected the sentence to properly read statement-based replication now. Thank you very much.Anonymoushttps://www.blogger.com/profile/10372091392084104297noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-38171177836792691292013-05-15T20:12:46.491-04:002013-05-15T20:12:46.491-04:00Great bug Daniël! I think sql_mode is a perfect wa...Great bug Daniël! I think sql_mode is a perfect way to handle it.Anonymoushttps://www.blogger.com/profile/10372091392084104297noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-39071664700442908662013-05-15T14:42:57.569-04:002013-05-15T14:42:57.569-04:00And I think your second sentence should read "...And I think your second sentence should read "In my opinion it is a much safer option than statement based replication."Daniël van Eedenhttps://www.blogger.com/profile/14757324605223498151noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-74392148597873925132013-05-15T14:41:33.544-04:002013-05-15T14:41:33.544-04:00You might find this bug interesting:
http://bugs.m...You might find this bug interesting:<br />http://bugs.mysql.com/bug.php?id=69223<br /><br />Tables without primary key are know to cause bad performance with mysql replication, InnoDB and with Galera.Daniël van Eedenhttps://www.blogger.com/profile/14757324605223498151noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-53499260305536192992013-05-15T09:58:01.822-04:002013-05-15T09:58:01.822-04:00Thanks for the advice Justin. I have done some rea...Thanks for the advice Justin. I have done some reading on 5.6 but was not aware of the slave hash algorithm or update batches. Those would be useful options for us in addition to setting the binary log row image to minimal and cutting down on disk space used.Anonymoushttps://www.blogger.com/profile/10372091392084104297noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-16032664851848362732013-05-15T09:47:38.887-04:002013-05-15T09:47:38.887-04:00Thanks Simon, I would use a feature like that. The...Thanks Simon, I would use a feature like that. The solution will likely continue to be user education on MySQL best practices.Anonymoushttps://www.blogger.com/profile/10372091392084104297noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-46129113563463437822013-05-14T18:35:32.790-04:002013-05-14T18:35:32.790-04:005.6 improves this. It can apply rows in batches a...5.6 improves this. It can apply rows in batches and use a HASH scan when an index is not available if the option is enabled:<br />http://dev.mysql.com/doc/refman/5.6/en/replication-options-slave.html#option_mysqld_slave-rows-search-algorithmsJustin Swanharthttps://www.blogger.com/profile/08193089637089861226noreply@blogger.comtag:blogger.com,1999:blog-6551499576288463009.post-21845004108135388332013-05-14T12:43:34.528-04:002013-05-14T12:43:34.528-04:00This is a perfect case to give a warning message f...This is a perfect case to give a warning message from MySQL for a situation like this, However any such warning needs to be rate-limited to ensure that the warnings do not fill your logs with information you've just been told about.Simon J Muddhttps://www.blogger.com/profile/02803644910012672552noreply@blogger.com