I’m making an INSERT SELECT.

 
Umbarger: Would I just have a table: vehicle_engine and each record belongsTo a vehicle, 1 = 1

Cuch: Sep 10 11:40:50 www mysqld: #011#011#011#011#011#011LEFT JOIN binaries b ON c.id=b.collection_id

Bolch: Sep 10 11:40:50 www mysqld: #011#011#011#011#011#011LEFT JOIN parts p ON b.id=p.binaryid

Dantin: Sep 10 11:40:50 www mysqld: #011#011#011#011#011#011LEFT JOIN groups g ON g.id = c.group_id

Trimble: Sep 10 11:40:50 www mysqld: #011#011#011#011#011#011WHERE c.filecheck = 3 AND c.group_id = 1000199

Blatz: Where the hell does that come from all of the sudden

Rumery: Bent0, find the beginning of one part, is there some “Unsafe statement written to the binary log” in there?

Dillmore: Djam90, from my limited knowledge of cars, they share engine types often, so you might find it’s a many to one mapping

Rumery: Bent0, #011 will be tab in your queries so those are only formatting – the actual message starts before that

Mcconchie: I do get a lot of these mysqld: 2015-09-10 11:28:38 688c92906700 InnoDB: current statement: DELETE c, b, p FROM collections c

Mullice: Cat /var/log/syslog grep Unsafe

Niimi: NeoThermicWork, looking at the CAP API the data is really stupid. Not normalised. I can only get the tech data per vehicle. I have to loop through every vehicle in my code to get all possible tech items

Rumery: Bent0, do you use replication?

Hulbert: Just changed binlog-format to MIXED

Mccollough: Grew to 10 mb in 2 mins

Rumery: So these are probably warnings about unsafe statements for mixed/statement format

Brumback: Sep 10 12:05:20 www mysqld: 2015-09-10 12:05:20 70d052518700 InnoDB: Error: unlock row could not find a 2 mode lock on the record

Koe: Sep 10 12:05:20 www mysqld: 2015-09-10 12:05:20 70d052518700 InnoDB: current statement: DELETE c, b, p FROM collections c

Jauhar: Sep 10 12:05:20 www mysqld: #011#011#011#011#011#011LEFT JOIN binaries b ON c.id=b.collection_id

Eisenberg: Do we use 0 as an ID? Imagine a client table. ID = 0 is a row with example content for a example client. ID’s 0 are actual implementations fo the real clients

Rumery: Bent0, whats your mysql version? I found a bug report for 5.1

Kassebaum: I know its possible, I just want to know if someone would advice against that practice and has a better alternative :

Migneault: Mysql Ver 15.1 Distrib 10.0.21-MariaDB, for debian-linux-gnu x86_64 using readline 5.2

Bodman: I did update php5-mysql last night

Rumery: Bent0, then try asking in #maria about “InnoDB: Error: unlock row could not find a 2 mode lock on the record” – maybe someone will respond

Hale: 0 is not allowed in auto_increment column

Rumery: Bent0, not totally and your first question was posting bad part of the error message :

Hird: Boleware, I know, but I can put it in myself as a example row.

Fornier: Example row = standart value row

Dittbrenner: P4trix: you can only insert it when there is no auto_increment

Rumery: Bent0, but google and read existing bug reports too, maybe something will help you

Kehs: I’m not sure what will happen if you make it auto_increment later

Grandi: Boleware, a ok. Then I can pick id = 1 as the standart value, and the following as values choosen by the client.

Cardinali: Disabled syslogging for now until I find a fix

Ober: Please help me with this query http://pastebin.pl/view/7ba20ce4

Curie: ArtisanIndia: ahaha i’ve seen this query before :p

Jone: ArtisanIndia: businesses.latitude

Josefy: Trying to get business in a range of user’s location

Nofziger: But I wrote it just now

Colombe: 6371 being earth radius :p

Corporal: ArtisanIndia: change it to businesses.latitude

Banahan: I’m making an INSERT SELECT and the inserted row are not in the same order as the select