Comments on: Getting rid of huge ibdata file, no dump required, part II https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii Blog by Shlomi Noach Sat, 20 Jul 2013 09:40:16 +0000 hourly 1 https://wordpress.org/?v=5.3.3 By: shlomi https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii/comment-page-1#comment-214904 Sat, 20 Jul 2013 09:40:16 +0000 https://shlomi-noach.github.io/blog/?p=4845#comment-214904 I agree. To be clear, I have not used the above approach on production. But looking back at past customer use cases, a couple of them could certainly benefit from it.

Thankfully by now there is little to no discussion about InnoDB vs myisam.

]]> By: Joop https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii/comment-page-1#comment-214902 Sat, 20 Jul 2013 08:55:43 +0000 https://shlomi-noach.github.io/blog/?p=4845#comment-214902 Think it’s an interesting approach that you’ve taken to resolve the ibdata file size, but I still the risks and overheads (investigation to see what might break) to achieve it are too high in a real-world environment. More so if the data is that of a financial nature. The safest is to do dump loads, or we can all just keep begging the MySQL developers to make a vacuum function to do background thread cleanup of ibdata files. This size issue of ibdata and idb files and the locking when doing optimizing is honestly one of the biggest flaws in MySQL, if they can fix this issue then MysQL could and would be so much more powerful than it is right now.

Keep up the blogging, some interesting reading here.

Joop

]]> By: shlomi https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii/comment-page-1#comment-214861 Fri, 19 Jul 2013 15:16:32 +0000 https://shlomi-noach.github.io/blog/?p=4845#comment-214861 @joop,

Thank you. I urge you to read fully, and follow link in last paragraph.

]]> By: Joop https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii/comment-page-1#comment-214809 Thu, 18 Jul 2013 14:54:23 +0000 https://shlomi-noach.github.io/blog/?p=4845#comment-214809 BAD idea….if you have constraints like delete cascades and you let the slave catch up your system is going to be out of sync. This is a REALLY BAD non-real world solution that should NEVER be used.

]]> By: shlomi https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii/comment-page-1#comment-97444 Sun, 03 Jun 2012 10:12:58 +0000 https://shlomi-noach.github.io/blog/?p=4845#comment-97444 @Daniel,
Thanks — fixed

]]> By: Daniël van Eeden https://shlomi-noach.github.io/blog/mysql/getting-rid-of-huge-ibdata-file-no-dump-required-part-ii/comment-page-1#comment-97441 Sun, 03 Jun 2012 09:53:08 +0000 https://shlomi-noach.github.io/blog/?p=4845#comment-97441 s/Altering to MySQL drops FK constraints/Altering to MyISAM drops FK constraints/ ?

]]>