[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: mico-2.3.0
From:       Steffen Hansen <stefh () mip ! sdu ! dk>
Date:       1999-09-09 21:20:33
[Download RAW message or body]

On Thu, 9 Sep 1999, Cristian Tibirna wrote:

> Hi Simon
> 
> I already got my answer to the question that I privately asked you
> last night (I didn't even check if you had time to answer).
> 
> YES! The CORBA (mico) base applications are working flawlessly
> immediately after your migration to mico-2.3.0. I was strongly impressed
> by the amount of work you've put into this. And all looks perfect from
> here.
> 
> And, as a bonus, I don't know if it's only me (some kind of
> autosuggestion) or it's real - can somebody please look for confirmation?
> - the office applications and konqueror are starting and running
> approximatively 5 times faster than before.

I can confirm that some bugs have disappeared. For example, the weird
racecondition where clients sometimes hang on the first invokation on an
object reference obtained from KActivator, has disappeared (i have hunted 
this bug for many hours without succes...). In general, the CORBA stuff
seems more robust now.
 
> Thanks many. And congrats, Simon

Yes, you have done a _great_ job here Simon! The recovery/restart code for
kded also seems to work, but there still are a few problems with it:

1) The client can't do errorhandling itself, since everything is
   catch(...)'ed.

2) If kded continues to crash, we have an infinite loop. This is exactly
   what happens if .kde/share/config/kregistry is garbled. Should i add a
   line to kded-server.cc that deletes kregistry if a crash is detected?

3) goto's are always interesting... :-)

greetings,
-- 
Steffen Hansen                            
email: stefh@mip.sdu.dk, stefh@imada.sdu.dk, hansen@kde.org 
URL:   http://www.mip.sdu.dk/~stefh       

[prev in list] [next in list] [prev in thread] [next in thread]