Clyatt: RQ: If it is script it will be much easier to set them elsewhere

Clyatt: RQ: You generate it with something don’t you?

Suiter: And then did some manual editing

Clyatt: RQ: Also why do you care about DEFAULTS that much and why do you bother to set them to high value?

Mcbrayer: 65534 is the uid of nobody and gid of nogroup

Koester: Uid=65534nobody gid=65534nogroup groups=65534nogroup

Clyatt: RQ: So when do you expect that to change?

Hemish: Or are you everyjust going to insert into the table and expect the defaults to affect the value of a row?

Tulk: Salle: when the user wants different defaults in his setup

Savas: Danblack: errr. what?

Clyatt: RQ: When do you expect Debian to change that 65534 to something else so you would need to reflect that in your scripts?

Jumper: Salle: I don’t expect Debian to change it. But the users might want different default user/group in their setup.

Clyatt: RQ: End users have no bussiness to ALTER tables. It is not their bussness to know the database structure at all

Wittnebel: The current version of that script has “CHANGE” as the default, and requires the user to actually change it before running the script

Brushwood: I thought I’d do it a bit smarter :

Sutler: But apparently it won’t work that way

Omalley: Salle: the “user” in this case is a mail server administrator

Shivers: You may be right, although that’s how the script worked for the past X years.

Atthowe: Hi. i’ve a table with a lot of row 4 thousand, it’s a lot? Now there’s a new requirement so i need to add a new column. What do you do in this case? Because there will be lots of rows with a null column.

Starrett: 4000 rows is a tiny table

Copenhaver: Why isn’t the index used when the query contains a varchar field?

Shill: Dkasll: Please paste your query, the EXPLAIN select., and the relevant SHOW CREATE TABLE/s in one pastebin with the sql formatted so it is readable

Capalongan: Boleware: as i was thinking, i was wrong

Cullop: RQ: If you insist, look into PREPARE / EXECUTE.

Gromley: Http://pastebin.com/vZxrfyfp

Suga: If SHOW ENGINE INNODB STATUS doesn’t list a LATEST DETECTED DEADLOCK, does that mean there has yet to be a deadlock?

Garib: Watmm: since the last restart, yes.

Santy: Watmm: are you worried about deadlocks for some reason?

Rumery: How can I get from “Using index; Using temporary; Using filesort” to “Using index; Using filesort” by ADDING one more distinct count? http://pastebin.com/nbsD0j60 — and the one NOT using temporary seems to be 1.7x slower

Rumery: I think I did have a similar situation some month ago, but did not save the queries. now it showed again

Rumery: If I can make it into a testcase

Lasseter: Why isn’t the index used when the query contains a varchar field? http://pastebin.com/vZxrfyfp

Rumery: Dkasll, because without it the index “covers” the query, but otherwise your table seems so small that using index in the second case would probably have no positive effect

Tervo: I’m using MySQL’s employees database, executing SELECT * FROM salaries where emp_no 1000 LIMIT 11, the query takes about 1.2 seconds to execute PMA says, the table only has about 2 million entries, is this expected result times?

Phanco: 2 millions is a lot of data to process

Montemarano: Is there an index on emp_no?

Erle: Boleware, right, but I’m asking whether that’s expected times? emp_no is a PK, I thought 2 millions was nothing for an int PK column

Giacolone: What does ‘explain select. ‘ show

Volper: Http://supa.me/ZJJOLw.png

Freshley: Delzer: Please paste your query, the EXPLAIN select., and the relevant SHOW CREATE TABLE/s in one pastebin with the sql formatted so it is readable