Inner join eliminate duplicates

Question

Schema:

Query:

 SELECT avto.Marka,SUM(OtrabotanoMachine_hours) as 'work hours',SUM(sec.Hours) as Downtime,CONVERT(group_concat(concat(prostoy.Prichina,'=',sec.Hours) separator ';') USING 'utf8') as 'downtime reasons',CONVERT(group_concat(concat(VipolnennayaRabota,'=',V_raboty) separator ';') USING 'utf8') as 'performed works'
FROM
jos_addRabotyAndProstoyMain main 
INNER JOIN avto ON main.Marka=avto.ID_Avto
INNER JOIN (
   SELECT id_fk, SUM(Hours) AS Hours, Prostoy
   FROM jos_addRabotyAndProstoySecond 
   GROUP BY id_fk,Prostoy) sec ON sec.id_fk = main.id 
INNER JOIN prostoy ON sec.Prostoy = prostoy.ID_Prosyoy
GROUP BY avto.Marka

Wrong result:

 | Marka   | work hours  |downtime| downtime reasons      | performed works|
 |JCB-90QX |      44     |    7   |lack of materials=2;   | work=51;       |
 |         |             |        |machine malfunction=1; | work=51;       |               
 |         |             |        |Others=3;              | work=51;       |
 |         |             |        |unscheduled repairs=1  | work=51;       |

The columns "work hours" and "performed works" contain duplicates. How eliminate duplicates?

I need result like this:

| Marka   | work hours  |downtime| downtime reasons      | performed works|
|JCB-90QX |      11     |    7   |lack of materials=2;   | work=51;       |
|         |             |        |machine malfunction=1; |                |               
|         |             |        |Others=3;              |                |
|         |             |        |unscheduled repairs=1  |                |

Update SQL Fiddle

Answer 1

I think the solution is to add a DISTINCT clause to the columns where you want to use unique data.

Try this (taken from your SQL Fiddle):

 SELECT avto.Marka,SUM(distinct OtrabotanoMachine_hours) as 'work hours',SUM(sec.Hours) as Downtime,CONVERT(group_concat(concat(prostoy.Prichina,'=',sec.Hours) separator ';') USING 'utf8') as 'downtime reasons',CONVERT(group_concat(distinct concat(VipolnennayaRabota,'=',V_raboty) separator ';') USING 'utf8') as 'performed works'
FROM
jos_addRabotyAndProstoyMain main 
INNER JOIN avto ON main.Marka=avto.ID_Avto
INNER JOIN (
   SELECT id_fk, SUM(Hours) AS Hours, Prostoy
   FROM jos_addRabotyAndProstoySecond 
   GROUP BY id_fk,Prostoy) sec ON sec.id_fk = main.id 
INNER JOIN prostoy ON sec.Prostoy = prostoy.ID_Prosyoy
GROUP BY avto.Marka

However, it is important to note that I am not 100% sure how the OtrabotanoMachine_hours is set up. If it is possible for there to be two legitimate separate entries with the same values, it will de-dupe them.