Re: Gatekeeper restarts

[Nour Omar <nouromar@xxxxxxxxxxxxx>]

Jan,

I'm wondering if this IRR feature is a replacement to tcpKeepAlive feature or the 2 have different functionality and purpose.  I was thinking about using IRR feature instead of tcpKeepAlive, then I thought what would happen if call is not disconnected because Release message is lost but the endpoint is still connected and responding to  IRR message?

let me/us know what you think.

Thanks

Jan Willamowius <jan@xxxxxxxxxxxxxx> wrote:

Nour,

Gnugk will terminate an active call if it doesn't receive the IRRs and
then after the TimeToLive the endpoint will be deregistered.

All changes (code and docs) are only in the CVS right now.


Nour Omar wrote:
> Jan,
> What would be the gatekeeper action if it does not receive IRR in the
specified frequency?
> I checked online manual and did not see your changes. Is it in CVS
too?
>
> Thanks
>
> Jan Willamowius wrote:
> I just added another feature that helps detect network problems:
> Now you can set the irrFrequency in ACF and check if both endpoint
keep
> sending them periodically. This is especially usefull when you don't
> want a short TimeToLive for all endpoint registrations (eg. because
> they are idle most of the time and ypu don't wan t the load), but you
> want to monitor those involved in a call very closely.
>
> Its in both the 2.0 and 2.2 CVS branch. Check the manual on the
> [CallTable] section.
>
>
> Regards,
> Jan
>
> PS: I hear the SourceForge does maintenance and parts of the CVS may
not
> be available today. Just try.
>
>
> Zygmuntowicz Michal wrote:
> > Make sure you have TimeToLive set to some reasonable
> > value in the config and your endpoints support ttl reregistration.
> >
> > ----- Original Message -----
> > From: "Guaina Montoya Alejandro"
> > Sent: Monday, August 08, 2005 5:57 PM
> >
> >
> > I just started using the gatekeeper for a videoconference project
> > demonstration over at my company.
> > Everything works fine, except for when I accidentally disconnected
the
> > Ethernet cable, so I had to restart the gatekeeper.
> > My question is, is there a way to tell all the endpoints to register
> > again? Or is that an endpoint problem, and therefore it doesn't
> concern
> > the gatekeeper?
> >
> > Alejandro Guaina
>
> --
> Jan Willamowius, jan@xxxxxxxxxxxxxx, http://www.gnugk.org/
>
>
> -------------------------------------------------------
> SF.Net email is Sponsored by the Better Software Conference & EXPO
> September 19-22, 2005 * San Francisco, CA * Development Lifecycle
Practices
> Agile & Plan-Driven Development * Managing Projects & Teams * Testing
& QA
> Security * Process Improvement & Measurement *
http://www.sqe.com/bsce5sf
> _______________________________________________________
>
> Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
> Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549
> Unsubscribe:
http://lists.sourceforge.net/lists/listinfo/openh323gk-users
> Homepage: http://www.gnugk.org/


--
Jan Willamowius, jan@xxxxxxxxxxxxxx, http://www.gnugk.org/


-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________________

Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549
Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users
Homepage: http://www.gnugk.org/