Am Montag 18 Juli 2016, 09:00:45 schrieb Lisandro Damián Nicanor Pérez Meyer:
> On domingo, 17 de julio de 2016 8:18:35 P. M. ART Lisandro Damián Nicanor 
> Pérez Meyer wrote:
> > As was posted [a couple of weeks ago], the latest version of KDE
> 
> This is of course KDEPIM :)

Of *course*. :-)

Updated my Sid box today, got KDEPIM and libs 16.04.3. Wanted to check whether
my favourite issue [0] had been fixed. =)

Created a fresh user on the box, logged in for the first time into Plasma5,
and started Kontact.

MySQL refused to start up properly (while tests 1..4 passed):

---8<---
Test 5:  ERROR
--------

MySQL server log contains errors.
Details: The MySQL server error log file '<a 
href="/home/chris/.local/share/akonadi/db_data/mysql.err">/home/chris/.local/share/akonadi/db_data/mysql.err</a>'
 contains errors.

File content of '/home/chris/.local/share/akonadi/db_data/mysql.err':
160719  9:09:39 [Note] Plugin 'FEDERATED' is disabled.
160719  9:09:39 InnoDB: The InnoDB memory heap is disabled
160719  9:09:39 InnoDB: Mutexes and rw_locks use GCC atomic builtins
160719  9:09:39 InnoDB: Compressed tables use zlib 1.2.8
160719  9:09:39 InnoDB: Using Linux native AIO
160719  9:09:39 InnoDB: Initializing buffer pool, size = 80.0M
160719  9:09:39 InnoDB: Completed initialization of buffer pool
160719  9:09:39 InnoDB: highest supported file format is Barracuda.
160719  9:09:39  InnoDB: Waiting for the background threads to start
160719  9:09:40 InnoDB: 5.5.46 started; log sequence number 1622641
160719  9:09:40 [Warning] Can't open and lock time zone table: Table 
'mysql.time_zone_leap_second' doesn't exist trying to live without them
160719  9:09:40 [ERROR] Can't open and lock privilege tables: Table 
'mysql.servers' doesn't exist
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'events_waits_current' has the wrong structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'events_waits_history' has the wrong structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'events_waits_history_long' has the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'setup_consumers' has 
the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'setup_instruments' 
has the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'setup_timers' has 
the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'performance_timers' 
has the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'threads' has the 
wrong structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'events_waits_summary_by_thread_by_event_name' has the 
wrong structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'events_waits_summary_by_instance' has the wrong structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'events_waits_summary_global_by_event_name' has the wrong 
structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'file_summary_by_event_name' has the wrong structure
160719  9:09:40 [ERROR] Native table 
'performance_schema'.'file_summary_by_instance' has the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'mutex_instances' has 
the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'rwlock_instances' 
has the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'cond_instances' has 
the wrong structure
160719  9:09:40 [ERROR] Native table 'performance_schema'.'file_instances' has 
the wrong structure
160719  9:09:40 [Note] /usr/sbin/mysqld: ready for connections.
Version: '5.5.46-0+deb8u1'  socket: '/tmp/akonadi-chris.raoZs2/mysql.socket'  
port: 0  (Debian)
160719  9:09:40 [Note] /usr/sbin/mysqld: Normal shutdown

160719  9:09:40  InnoDB: Starting shutdown...
160719  9:09:42  InnoDB: Shutdown completed; log sequence number 1622651
160719  9:09:42 [Note] /usr/sbin/mysqld: Shutdown complete
---8<---

>From what I found in the reports, this looks like [1].

I have checked that

* there is enough free disk space (~7GB free)
* the Akonadi directory structure under ~/.local/share/akonadi has the right 
permissions
  (this is not a migrated DB, but freshly created, as noted above)

So I thought this might be a wrong default setting with the table types, as 
Akonadi requires
InnoDB, as noted in [1]. But then, the log above shows that InnoDB is used.

As I am running the tests inside a virtual box, I assumed I might have had too 
little memory
reserved. Cleared the users' home, increased virtual machine mem from 4GB to 
8GB and started
over.

Alas, same thing happened. A migration might eat up truckloads of mem, but a 
fresh install
with no IMAP account yet -- just the MySQL DB -- I think 8GB of mem should 
suffice.

After quitting the email account wizard and Kontact alltogether, and starting 
Kontact anew,
no error dialog was shown. However, 'akonadictl status' keeps telling that the 
service is
down.

Trying the hint provided in [2] (deleting the db_data Akonadi directory and 
re-creating it
with mysql_install_db) I found that the MySQL libs versions 5.5 were installed, 
and the
MySQL server itself was not. mysql_install_db complained about resolveip not 
being able
to resolve the host name. Missing the MySQL server, resolveip was not even 
installed.

After installing the MySQL 5.6 server, I re-ran

'mysql_install_db --datadir=$HOME/.local/share/akonadi/db_data/ --basedir=/usr'

just to see what will happen.

'akonadictl start' was then able to start up, Akonadiconsole showed the agents 
running
and Kontact would come up.

So, I logged off from Plasma5, cleared out the users' home once more and again 
started
over, reducing VM mem to 4GB again.

This time -- whoa! -- no errors while starting up Akonadi. I was then able to 
sync the
IMAP folders down to the box (I always use the offline option).

Unsure whether the full MySQL 5.6 server is needed, but when I upgraded from 
Stretch
to Sid, obviously, there was missing some dependency which would allow me to 
set up
a fresh Akonadi DB.

While syncing the IMAP folders, the process got stuck at 100% at one folder, as 
in [3].
Trying to resolve this now...


Kind regards,

        Christian


[0] https://bugs.kde.org/show_bug.cgi?id=283682
[1] https://bugs.kde.org/show_bug.cgi?id=333736
[2] https://bbs.archlinux.org/viewtopic.php?id=78358
[3] https://bugs.kde.org/show_bug.cgi?id=274799

-- 
kernel concepts GmbH       Tel: +49-271-771091-11
Sieghuetter Hauptweg 48
D-57072 Siegen
http://www.kernelconcepts.de/
Geschäftsführer: Nils Faerber, Ole Reinhardt
HR Siegen, HR B 9613

Reply via email to