Re: Multisync: patch and question

Gonzalo Diethelm wrote:
> On Fri, 2005-03-11 at 17:57 -0300, Gonzalo Diethelm wrote:
>
>>>What you definitely should do is set the last parameter to a GError * so
>>>you can see what evo2 is complaining about :)
>>
>>I did that, and got:
>>
>>CALENDAR error was Unknown error
>>
>>Unluckily, not very useful...
>>
>>
>>>I dont think thats the cause. Its the same here.
>>
>>Ok, it was a wild guess anyway. Will try with 0.9.
>
>
> I have another question related to this: how do you effectively debug
> Mulsitync 0.8? I tried strace-ing the process (and all the forked
> processes as well), hoping to find an open call that would open the
> calendar.ics-msyncidX.db file, but could not find it in the traces at
> all. I guess this is because Multisync is using some kind of
> out-of-process invocation (CORBA?) to query Evolution's DB. If this is
> the case, how can I trace the opening, reading and closing of the DB?
>

there is no mechanism besides export MULTISYNC_DEBUG=1
The absence of good debugging support is also a reason for abondining
multisync 0.8X, its almost impossible to find bugs.

yes, evo2's api uses rpc to communicate with the e-d-s. you can trace
the opening etc by killing the evolution data server:
/usr/lib/evolution/2.0/killev and then start it manually: strace
evolution-data-server.

> If this is all completely off mark, how should I go about finding out
> exactly what happens that makes Multisync abort with the "Unknown error"
> message?
>

Try if this helps: add a entry to the calendar before syncing.

> Regards,
>
>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)

iD8DBQFCQGQRq9z7v9k9UakRAn6tAJ40uaQiphontRmHW6TFldDNv/a/tACfVxQ7
l4/T3eqjbl7Ky9e2oGiGD0Q=
=M4KF
-----END PGP SIGNATURE-----