Bug 333239

Summary: akonadi refuses to start after disk was full
Product: [Frameworks and Libraries] Akonadi Reporter: Christian Boltz <kde-bugs>
Component: serverAssignee: kdepim bugs <kdepim-bugs>
Status: RESOLVED UNMAINTAINED    
Severity: normal CC: dvratil, rdieter
Priority: NOR    
Version: 4.12   
Target Milestone: ---   
Platform: openSUSE   
OS: Linux   
Latest Commit: Version Fixed In:
Sentry Crash Report:

Description Christian Boltz 2014-04-09 18:21:13 UTC 
My disk ran full while akonadi and kmail were running.

The symptoms I noticed were that kmail wasn't able to display mail content anymore - I'd guess because it tried to write them to a tempfile on the full disk first.

After freeing some disk space again, akonadi refuses to start.

The debug output shows that it tries to start a mysql instance (which fails), however I had it configured to use the system-wide mysql. 

In other words: akonadi destroyed its config file when the disk was full!

Reproducible: Didn't try

Steps to Reproduce:
1. start akonadi and kmail, ideally with akonadi configured to use the system-wide mysql
2. let another process fill up your disk (if you don't have a better idea, dd if=/dev/zero of=~/fill-my-disk)
3. quit kmail and akonadi
4. (try to) restart akonadi


Expected Results:  
The config file should not be destroyed when the disk is full.
Comment 1 Christian Boltz 2014-04-09 18:33:05 UTC 
Just FYI: restoring ~/.config/akonadi/ from the backup fixes this issue.

The "broken" ~/.config/akonadi has lots of agent_config_akonadi*.dat files with 0 bytes. I'd guess akonadiserverrc was also 0 bytes, but can't verify it because it was (probably) overwritten by akonadiconsole.
Comment 2 Denis Kurz 2016-09-24 20:41:04 UTC 
This bug has only been reported for versions older than KDEPIM 4.14 (at most akonadi-1.3). Can anyone tell if this bug still present?

If noone confirms this bug for a recent version of akonadi (part of KDE Applications 15.08 or later), it gets closed in about three months.
Comment 3 Denis Kurz 2017-01-07 21:49:58 UTC 
Just as announced in my last comment, I close this bug. If you encounter it again in a recent version (at least 5.0 aka 15.08), please open a new one unless it already exists. Thank you for all your input.