Akonadi dsaster

[Pablo Sanchez]

[ Comments below, in-line ]


On Tue, 11 Oct 2016 14:54:07 +0200, Martin Steigerwald wrote:

> Am Dienstag, 11. Oktober 2016, 11:54:45 CEST schrieb Diederik de
> Haas:
>> On dinsdag 11 oktober 2016 11:03:16 CEST Martin Steigerwald wrote:

>>
>> At some (near?) point in time I will switch to PostgreSQL on my
>> laptop too and my guess is that there's a real good chance that it
>> would improve Akonadi's working considerably. There is one downside
>> to using PostgreSQL as Dan explained in [1]. But once you've
>> figured out a proper procedure for it [2], it's actually straight
>> forward
>
> Thank you for the links. I am aware of Dan using PostgreSQL. I used
> it once myself. Even system-wide for a while, but also user-wide.
> With PostgreSQL I dislike that on major upgrade you need to migrate
> your database AFAIK. I see the links are exactly about this. Thank
> you for the pointer!
>

Hi everyone,

I'm a database engineer.  The last time I looked at what was going on 
with Akonadi's use of MySQL, I saw an incessant amount of SQL calls per 
second.  All the data was in cache so from the O/S perspective, it 
manifested itself as high CPU utilization.

Dan put in a hack to alleviate *some* of the pressure.  But it was still 
not great.  The thinking was to provide some relief before a major 
refactoring.  I don't know what happened.

I'm more than happy to spin up a VM and do another round of 
investigation.  But Dan has to be open.  I (like everyone else) am a bit 
busy with work so I want to make sure my efforts are not wasted.  :)

Cheers,

--
Pablo Sanchez - Blueoak Database Engineering, Inc
Ph:    819.459.1926        iNum:  883.5100.0990.1054