https://shlomi-noach.github.io/blog/mysql/state-of-inndb-online-ddl-in-mysql-5-6-9-rc-good-news-included Blog by Shlomi Noach Fri, 25 Jan 2013 00:20:31 +0000 hourly 1 https://wordpress.org/?v=5.3.3 https://shlomi-noach.github.io/blog/mysql/state-of-inndb-online-ddl-in-mysql-5-6-9-rc-good-news-included/comment-page-1#comment-166928 Fri, 25 Jan 2013 00:20:31 +0000 https://shlomi-noach.github.io/blog/?p=5882#comment-166928 Also, ALTER PK statements now take place in @@tmpdir. That could require a rearchitecture for many people.

]]> https://shlomi-noach.github.io/blog/mysql/state-of-inndb-online-ddl-in-mysql-5-6-9-rc-good-news-included/comment-page-1#comment-166920 Thu, 24 Jan 2013 23:56:03 +0000 https://shlomi-noach.github.io/blog/?p=5882#comment-166920 Unfortunately, PK are not an online schema change. I wonder if this will be fixed by .11? Never the less I am very impressed with all the efforts from Oracle and others!

]]> https://shlomi-noach.github.io/blog/mysql/state-of-inndb-online-ddl-in-mysql-5-6-9-rc-good-news-included/comment-page-1#comment-150800 Thu, 20 Dec 2012 06:05:55 +0000 https://shlomi-noach.github.io/blog/?p=5882#comment-150800 Hi Robert,
Yes, I suppose that would make sense best.

]]> https://shlomi-noach.github.io/blog/mysql/state-of-inndb-online-ddl-in-mysql-5-6-9-rc-good-news-included/comment-page-1#comment-150015 Tue, 18 Dec 2012 15:39:56 +0000 https://shlomi-noach.github.io/blog/?p=5882#comment-150015 Hi Shlomi! Thanks for testing this. Speaking from the perspective of Tungsten (the other parallel replication) it is correct that the ALTER TABLE will be serialized. It’s dangerous to execute DDL statements out of order automatically. Assuming this is a rare operation another approach is to execute directly on each affected server using an unlogged command so that it does not even go into the binlog.

]]>