<html><body>
<p><tt>Hi Alex,<br>
<br>
I think it's a convenient way to manage the logfiles, I</tt><br>
<tt>didn't know about it. Then, maybe you can send a patch to </tt><br>
<tt>the list to modify omniNames or, if you want, I can take</tt><br>
<tt>a look at your code and try to integrate it.</tt><br>
<br>
<tt>Thanks,</tt><br>
<tt>Jorge<br>
</tt><br>
<tt>Alex Tingle <alex.tingle@bronermetals.com> escribió el 07/09/2005 13:25:01:<br>
<br>
> Hi Jorge,<br>
> <br>
> omniEvents used to have exactly the same behaviour, and I too was <br>
> dissatisfied with it. I changed omniEvents to treat its "-start" <br>
> parameter (actually -p) like this:<br>
> <br>
> If there is no logfile then create one.<br>
> <br>
> If there is a logfile and the port in the file matches the start <br>
> parameter, then start up as normal: ignore the parameter.<br>
> <br>
> If there is a logfile and the port in the file DOES NOT match the <br>
> start parameter, then quit with an error.<br>
> <br>
> I feel that this gets the balance right between convenience and <br>
> preserving the way it works now.<br>
> <br>
> -Alex<br>
> <br>
> --<br>
> Dammit Jim, I'm a programmer... not a mind reader!<br>
> On 7 Sep 2005, at 12:01, jorgefm@cirsa.com wrote:<br>
> <br>
> > Well, the only thing I dislike from the current behaviour<br>
> > is the error that omniNames generates now. I prefer to<br>
> > have a known initial situation in my embedded device when<br>
> > he boots, and an option, is to recreate the log file every<br>
> > time.<br>
> ><br>
> > I would prefer that if the log is present omniNames would use<br>
> > it, and if the log doesn't exists omniNames would create it,<br>
> > avoiding the need of the parameters. But my solution is more easy<br>
> > to integrate in the existing code.<br>
> ><br>
> >> than once? Once you've created the database, why do you want to<br>
> >> re-create it?<br>
> >><br>
> ><br>
> > Alex Tingle <alex.tingle@bronermetals.com> escribió el 07/09/2005<br>
> > 12:39:18:<br>
> ><br>
> >> Um, why do you want to use omniNames with the -start parameter more<br>
> >> than once? Once you've created the database, why do you want to<br>
> >> re-create it?<br>
> >><br>
> >> -Alex<br>
> >><br>
> >> --<br>
> >> Dammit Jim, I'm a programmer... not a mind reader!<br>
> >> On 7 Sep 2005, at 11:19, jorgefm@cirsa.com wrote:<br>
> >><br>
> >>> Hi all,<br>
> >>><br>
> >>> I've made a little patch to omniNames from the omniORB 4.0.6, that I<br>
> >>> would<br>
> >>> like to be included in the main branch if it is useful for you.<br>
> >>><br>
> >>> The patch tries to address the problem that I've always that I try<br>
> > to<br>
> >>> execute the omniNames. I've change the -start parameter behaviour.<br>
> >>><br>
> >>> 1) When the -start is used, I delete the old log files instead of<br>
> >>> generating<br>
> >>> an error, but this can be controlled using the 'removelogs'<br>
> > constructor<br>
> >>> argument.<br>
> >>><br>
> >>> 2) The permission given to the logs files are RW for all (owner,<br>
> > group<br>
> >>> and others)<br>
> >>><br>
> >>> Thanks for your comments,<br>
> >>> Jorge<br>
> >>><br>
> >>> (Ver archivo anexo: omniNames-4.0.6-patch)<br>
> >>><br>
> >>><br>
> > <omniNames-4.0.6-patch>_______________________________________________<br>
> >>> omniORB-list mailing list<br>
> >>> omniORB-list@omniorb-support.com<br>
> >>> <a href="http://www.omniorb-support.com/mailman/listinfo/omniorb-list">http://www.omniorb-support.com/mailman/listinfo/omniorb-list</a><br>
> > <<a href="http://www.omniorb-support.com/mailman/listinfo/omniorb-list">http://www.omniorb-support.com/mailman/listinfo/omniorb-list</a>><br>
> >><br>
> ><br>
> ><br>
> <br>
</tt></body></html>