kmail sigh

Sandro Knauß sknauss at kde.org
Mon Jan 29 11:45:59 GMT 2018


Hey,

> Try converting kmail to use postgresql, its a lot faster.  It doesn't
> eliminate akonadi issues like duplicates but i seem to get a hell of lot
> less of them and the "retrieving folders..." issue seems to have
> disappeared.  If you have to do the "akonadictl fsck/vacuum", you'll find
> that is so fast compared to running it using mysql

I do not believe that the switch by itself makes things a lot faster ( So you 
have sum up the time the database processes are consuming resources?), but 
with a switch to postgres you get rid of all broken elements in the database 
(reported by akonadictl fsck) and so Akonadi can now sync properly afterwards 
and do not run every time into the slow error correction paths. That may be 
the main reason, why it is that damn more faster than before. And from a 
political point of view I can support the switch to postgres too :D 

So it may be not real the reason why it get faster, but in the end it doesn't 
matter what make kmail faster :)

Did anyone have made a measurement for the different back ends?

hefee



More information about the kdepim-users mailing list