* autofs segfault at startup on x86_64
@ 2006-01-03 10:49 Guillaume Rousse
2006-01-03 11:30 ` Ian Kent
0 siblings, 1 reply; 4+ messages in thread
From: Guillaume Rousse @ 2006-01-03 10:49 UTC (permalink / raw)
To: autofs
I've been instructed to report this here instead of kernel bugzilla:
http://bugzilla.kernel.org/show_bug.cgi?id=5561
--
Trucks that overturn on the highway are alway filled with something sticky
-- Murphy's Driving Laws n°3
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: autofs segfault at startup on x86_64
2006-01-03 10:49 autofs segfault at startup on x86_64 Guillaume Rousse
@ 2006-01-03 11:30 ` Ian Kent
2006-01-03 12:48 ` Guillaume Rousse
0 siblings, 1 reply; 4+ messages in thread
From: Ian Kent @ 2006-01-03 11:30 UTC (permalink / raw)
To: Guillaume Rousse; +Cc: autofs
On Tue, 2006-01-03 at 11:49 +0100, Guillaume Rousse wrote:
> I've been instructed to report this here instead of kernel bugzilla:
> http://bugzilla.kernel.org/show_bug.cgi?id=5561
>
Hi Guillaume,
Yes, I'll follow up on this.
Perhaps a simple patch for configure would be a good workaround while I
check out what's required. It appears you have spent a fair bit of time
on this already but it would help if you could give me a brief
description of your understanding of the problem and what you have found
out so far.
Ian
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: autofs segfault at startup on x86_64
2006-01-03 11:30 ` Ian Kent
@ 2006-01-03 12:48 ` Guillaume Rousse
2006-01-04 16:14 ` Ian Kent
0 siblings, 1 reply; 4+ messages in thread
From: Guillaume Rousse @ 2006-01-03 12:48 UTC (permalink / raw)
To: autofs
Ian Kent wrote:
> On Tue, 2006-01-03 at 11:49 +0100, Guillaume Rousse wrote:
>
>>I've been instructed to report this here instead of kernel bugzilla:
>>http://bugzilla.kernel.org/show_bug.cgi?id=5561
>>
>
>
> Hi Guillaume,
>
> Yes, I'll follow up on this.
>
> Perhaps a simple patch for configure would be a good workaround while I
> check out what's required. It appears you have spent a fair bit of time
> on this already but it would help if you could give me a brief
> description of your understanding of the problem and what you have found
> out so far.
Autofs requires to be compiled with -DLDAP_DEPRECATED on x86_64
platforms when linked against openldap >= 2.3, otherwise it
automatically triggers a segfault in openldap library during
initialisation (more exactly, in ldap_set_option() call from
do_connect() in lookup_ldap.c).
OpenLDAP maintainers suggest to either use this flags automatically
during compilation, or to change autofs code to use newest API.
--
No flight ever leaves on time unless you are running late and need the
delay to make the flight
-- Murphy's Laws for Frequent Flyers n°1
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: autofs segfault at startup on x86_64
2006-01-03 12:48 ` Guillaume Rousse
@ 2006-01-04 16:14 ` Ian Kent
0 siblings, 0 replies; 4+ messages in thread
From: Ian Kent @ 2006-01-04 16:14 UTC (permalink / raw)
To: Guillaume Rousse; +Cc: autofs
On Tue, 3 Jan 2006, Guillaume Rousse wrote:
> Ian Kent wrote:
> > On Tue, 2006-01-03 at 11:49 +0100, Guillaume Rousse wrote:
> >
> >>I've been instructed to report this here instead of kernel bugzilla:
> >>http://bugzilla.kernel.org/show_bug.cgi?id=5561
> >>
> >
> >
> > Hi Guillaume,
> >
> > Yes, I'll follow up on this.
> >
> > Perhaps a simple patch for configure would be a good workaround while I
> > check out what's required. It appears you have spent a fair bit of time
> > on this already but it would help if you could give me a brief
> > description of your understanding of the problem and what you have found
> > out so far.
> Autofs requires to be compiled with -DLDAP_DEPRECATED on x86_64
> platforms when linked against openldap >= 2.3, otherwise it
> automatically triggers a segfault in openldap library during
> initialisation (more exactly, in ldap_set_option() call from
> do_connect() in lookup_ldap.c).
Thanks for the info.
I chaecked the thread you reffered to about this and I think that a
configure patch for autofs-4.1.4 would be adequate for that version.
People using older versions could backport it.
The only question I have to answer is whether it would be OK to use the
-DLDAP_DEPRECATED on all platforms for openldap >= 2.3 or perhaps > 2.2?
>
> OpenLDAP maintainers suggest to either use this flags automatically
> during compilation, or to change autofs code to use newest API.
Yes. The LDAP code is badly in need of work. I have a patch in the queue
which greatly improves the code and upgrading the API calls for 4.1.5 and
beyond will probably prove to be a good idea. I'll also need to check how
far back the newer API calls have been available to make sure that people
with older OpenLDAP versions don't break. And there's the RedHat Directory
Server in Fedora to support as well now.
I'm not very familiar with the LDAP API, I just use what I need to get the
job done in autofs so any information that might be volunteered by people
more familiar would be greatly appreciated.
Ian
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2006-01-04 16:14 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-01-03 10:49 autofs segfault at startup on x86_64 Guillaume Rousse
2006-01-03 11:30 ` Ian Kent
2006-01-03 12:48 ` Guillaume Rousse
2006-01-04 16:14 ` Ian Kent
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.