3.0.2 / rlm_rest - segmentation fault when "Opening additional connection"

Chaigneau, Nicolas nicolas.chaigneau at capgemini.com
Fri Mar 28 14:18:42 CET 2014


Woops. Yes I did comment out "connect_uri" :)

(in my defense, the configuration example said I could)

Thanks for the quick update.


I've noticed that the module is not in the "stable" list, but it's fine. For the moment I'm just messing around to see if it can fulfill my needs.
If it can reach a stable point soon, even better!



Regards,
Nicolas.



> -----Message d'origine-----
> De : freeradius-users-
> bounces+nicolas.chaigneau=capgemini.com at lists.freeradius.org
> [mailto:freeradius-users-
> bounces+nicolas.chaigneau=capgemini.com at lists.freeradius.org] De la
> part de Arran Cudbard-Bell
> Envoyé : vendredi 28 mars 2014 12:08
> À : FreeRadius users mailing list
> Objet : Re: 3.0.2 / rlm_rest - segmentation fault when "Opening
> additional connection"
> 
> 
> On 28 Mar 2014, at 10:24, Chaigneau, Nicolas
> <nicolas.chaigneau at capgemini.com> wrote:
> 
> > Hello,
> >
> >
> > I'm trying to use rlm_rest module, with no luck so far. I have a
> segmentation fault when it needs to add a connection to the pool.
> > Any idea what the issue could be ?
> 
> Yeah, you've commented out connect_uri! Bad user, bad!
> 
> I'll push a fix.
> 
> You can just change:
> 
> 	if (!*inst->connect_uri) {
> 		ERROR("rlm_rest (%s): Skipping pre-connect, connect_uri not
> specified", inst->xlat_name);
> 		return candle;
> 	}
> 
> in rest.c:
> 	if (!inst->connect_uri) {
> 		ERROR("rlm_rest (%s): Skipping pre-connect, connect_uri not
> specified", inst->xlat_name);
> 		return candle;
> 	}
> 
> Or as a work around, just set the connect URI to any URL on the server
> you're using,
> that's the recommended configuration anyway, as it means there's no add
> latency
> establishing the TCP connection the first time a handle is used to
> service a request.
> 
> Thanks for the bug report.
> 
> Note: Before 3.0.3 (unreleased) rlm_rest is marked as an unstable
> module.
> 
> It was the large bit chunk of C code I wrote, and although the code
> organisation and general style
> is ok, corner cases aren't always dealt with gracefully.
> 
> When it was written, it involved changes to the FreeRADIUS core API to
> support new features,
> and unfortunately that API changed considerably between the time the
> module was written
> and FreeRADIUS 3.0.0 being released. That's part of the reason why
> there have been quite a few
> issues with it.
> 
> -Arran
> 
> Arran Cudbard-Bell <a.cudbardb at freeradius.org>
> FreeRADIUS Development Team
> 
> FD31 3077 42EC 7FCD 32FE 5EE2 56CF 27F9 30A8 CAA2

This message contains information that may be privileged or confidential and is the property of the Capgemini Group. It is intended only for the person to whom it is addressed. If you are not the intended recipient, you are not authorized to read, print, retain, copy, disseminate, distribute, or use this message or any part thereof. If you receive this message in error, please notify the sender immediately and delete all copies of this message.



More information about the Freeradius-Users mailing list