Stieglitz: Brittman, for a wordpress installation within a magento site . WP db is not being backed up.
Brittman: Laurentide: you probably want backup automation on the mysql server directly, there are existing tools for this
Saborido: Brittman: so my question is whether a query insert or update or delete can take more than 1 us?
Ballena: Agreed. i will be p***ing the suggestion along to the folks who implement decisions =
Griner: Xor1: You are probably asking the wrong question.
Greminger: Hi All, question about MySQL Select statement and how it reads columns into buffer memory?
Mador: Does MySQL 5.6 read all columns from disk to buffer/memory even if I select a few columns in the Select statement?
Carroway: Any help would be greatly appreciated
Trebbe: Silent_Ninja: It does that mainly by table and by index, with some exceptions for blob types.
Brittman: Silent_Ninja: http://mysqlserverteam.com/externally-stored-fields-in-innodb/
Nadal: Xgc: hmm. Let’s say that I have a trigger for insert/update/delete for a table. After all these ops, I store unix_timestampcurtime6 in some table. Is it possible that 2 separate trigger event will result in same timestamp?
Brittman: Silent_Ninja: the page is read, things stored off page and not selected are not
Abramek: Thanks all, so MySQL reads the full rows even though if I selected only 2 or 3 columns and for blogs, TEXT they maybe off external tables?
Knetsch: Silent_Ninja: Not in the case an index can be used. See the concept of a covering index.
Johnting: Thanks all, I guess having a good table structure with optimized column Data types is critical. Need to fix a column with dat type TEXT. great support channel!
Brittman: Silent_Ninja: full pages are read/written by innodb 16KiB
Brittman: Silent_Ninja: spend some time with that link i sent you
Alecca: Will do thanks again!
Brittman: Other wonderful posts: http://blog.jcole.us/2013/01/03/the-basics-of-innodb-space-file-layout/
Brittman: Xgc: wrt whether a covering index is used and in the buffer pool?
Schussler: Brittman: Right. The base table and storage may not be involved if an index is sufficient. We may be reading an index only and not the table.
Proa: Silent_Ninja: See http://dev.mysql.com/doc/refman/5.6/en/internal-temporary-tables.html
Brittman: Also something to think about
Gritman: Yep will do, thanks all!
Brittman: Bah gcs.fc_* aren’t dynamic.
Brittman: Warning WSREP: Setting parameter ‘gcs.fc_debug’ to ‘1000’ failed: Setting ‘gcs.fc_debug’ to ‘1000’ failed: 22 Invalid argument
Forslund: Brittman: :- will devs ever learn that restarting isn’t a desireable option
Brittman: Danblack: no worries just use another node.
Brittman: Danblack: any guess why an idle node would send flow control? ;
Izsak: All temporary tables are in-memory!
Izsak: Just with slightly different swap-out-to-disk rules
Strissel: Brittman: that one stumps me. not set read only or something and has a large queue because of it?
Brittman: Danblack: no connections what-so-ever
Brittman: I’m going to profile it with perf if i can time it right
Andreessen: Hi, http://pastebin.com/6wMV6WcV , I don’t know why refuses to connect
Starrs: Gcosmin: can you connect to mysql using the mysql client, if so the question might be better posed in a Java channel
Lamarta: Lost_soul: I can use phpmyadmin and it works fine
Hayduk: Gcosmin: I would ask your question in a Java channel since connections seem to be working fine otherwise
Brittman: Maybe one uses localhost unix socket and the other tcp eg 127.0.0.1
Copier: What makes innodb buffer pool dirty pages do hills and valleys in flushing? I mean, I know I can tune innodb_max_dirty_pages_pct, but why does the wo***oad change so drastically when I do? Some values provide stable numbers, other values provide dirty pages steadily increasing, then flushing to 0% dirty.