[SSSD] [PATCHES] CONFIGURE: Enhance detection of pam

Jakub Hrozek jhrozek at redhat.com
Tue Jun 3 11:55:55 UTC 2014


On Tue, Jun 03, 2014 at 10:55:05AM +0200, Jakub Hrozek wrote:
> On Tue, Jun 03, 2014 at 09:50:27AM +0200, Sumit Bose wrote:
> > On Tue, Jun 03, 2014 at 09:37:12AM +0200, Lukas Slebodnik wrote:
> > > On (03/06/14 09:35), Sumit Bose wrote:
> > > >On Tue, Jun 03, 2014 at 09:19:09AM +0200, Lukas Slebodnik wrote:
> > > >> On (03/06/14 09:06), Sumit Bose wrote:
> > > >> >On Tue, Jun 03, 2014 at 08:15:36AM +0200, Lukas Slebodnik wrote:
> > > >> >> On (02/06/14 23:19), Sumit Bose wrote:
> > > >> >> >On Sat, May 31, 2014 at 04:59:44PM +0200, Lukas Slebodnik wrote:
> > > >> >> >> ehlo,
> > > >> >> >> 
> > > >> >> >> I think that the attached patches are self descriptive.
> > > >> >> >> 
> > > >> >> >> LS
> > > >> >> >
> > > >> >> >Patches apply and build fine. I've run build on a few platforms without
> > > >> >> >any issue, but I had none with openpam at hand.
> > > >> >> >
> > > >> >> >I only have one comment on the last patch. getlogin() is not thread safe
> > > >> >> It is
> > > >> >> https://github.com/freebsd/freebsd/blob/master/lib/libc/gen/getlogin.c
> > > >> >>     lines 72:82
> > > >> >
> > > >> >There are locks, but the result is still in a static variable (line 59),
> > > >> >i.e. in the admittedly rare case when two threads do authentication in
> > > >> >parallel for different users one thread might override the result of the
> > > >> >other.
> > > >> >
> > > >> 
> > > >> Attaching version withou getlogin. It will require more time to find proper
> > > >> solution and I do not want to block other patches in this patchset.
> > > >
> > > >ok, you have my ACK here. I just wonder if you want to add a replacement
> > > >for pam_modutil_getlogin() which just returns NULL for the time being to
> > > >make SSSD compile on FreeBSD? Since the return value is only used for
> > > >log messages I think this will do no harm.
> > > >
> > > In future, patchset does not solve all problems :-)
> > 
> > ok, ACK.
> > 
> > bye,
> > Sumit
> 
> Pushed to master:
>     683e1f67d08be7165ea456d4594c4c8a4eddc9b3
>     78eae10aa035bc0e0996189b9d11c8db4904a98d
>     9e17b0516d2b640d34cbb50933bd065e65bfab9d
>     d2d21d45a88d0869f860e215c2cf946f032818ca
>     5385172799f6dea59fe1c0ef8f482db59bfc9c5c
>     81d6673764c4e2f635482be1efd52eba3ab5a27f

Also pushed to sssd-1-11:
    0690f3222a69607843433e6ad0870af95193801e
    368f03b2c3c63e82f8f9073f851143ff270d62a6
    b6d7e01b4b76cdc72cde36e8cc7c7216fd3cdb6b
    093285313c7e3d69069ea0d4637f7fee00fb2a2e
    1659171ed9585201df90003559884e070bd9cb69
    13e0c411b3106663399ca9fcd79c55ae10276873
    eb774aff170b3ea671dacaf50aee5595cce1ae9a
    4d2d5690159a0ccc281e53951981a5c0f1641eee
    6f895da367934fe44e101d8bc3d7dcee11034f3b



More information about the sssd-devel mailing list