Go to the FreeLists Home Page Home Signup Help Login
 



[openbeos] || [Date Prev] [03-2005 Date Index] [Date Next] || [Thread Prev] [03-2005 Thread Index] [Thread Next]

[openbeos] Re: CIA bot

  • From: "Axel Dörfler" <axeld@xxxxxxxxxxxxxxxx>
  • To: openbeos@xxxxxxxxxxxxx
  • Date: Mon, 21 Mar 2005 19:08:59 +0100 CET
Niels Reedijk <niels.reedijk@xxxxxxxxx> wrote:
> > > I've no idea, but I wouldn't mind if we find another method of 
> > > using it
> > > anyway, as it considerably slows down the commit. Without it, 
> > > there is
> > > almost no wait when commiting via Subversion (unlike CVS, that's 
> > > why it
> > > suddenly matters to me :-)).
> I don't really know if this is true. As you can see in the post-
> commit
> hooks, the ciabot_svn.py is called with an ampersand after it, so 
> this
> means that it would run asynchronously

Maybe the connection is closed only if all apps are terminated. A 
redirection of the input/output to /dev/null could bring help in that 
case.

But I can only talk from my experiences: directly after you enabled the 
CIA bot again, commits were very slow. Even ViewCVS via the web would 
already show all changes, but the "svn commit" command would still wait 
for something. On the next day, commits were suddenly fast again, and I 
thought "oh, maybe it was only a temporary thing with CIA" - but when I 
visited the CIA bot page, I learned that those fast commits weren't 
taken into account.
That looks pretty obvious to me, but it's not a proof either.

Bye,
   Axel.






[ Home | Signup | Help | Login | Archives | Lists ]

All trademarks and copyrights within the FreeLists archives are owned by their respective owners.
Everything else ©2007 Avenir Technologies, LLC.