Hi,
one of the issues reported in https://bugzilla.redhat.com/show_bug.cgi?id=975001 is a core dump in the PAC responder. I thought that I can find missing results by checking for ENOENT which is wrong in this case. This patch properly checks the number of results.
bye, Sumit
On Tue, Jun 18, 2013 at 05:30:09PM +0200, Sumit Bose wrote:
Hi,
one of the issues reported in https://bugzilla.redhat.com/show_bug.cgi?id=975001 is a core dump in the PAC responder. I thought that I can find missing results by checking for ENOENT which is wrong in this case. This patch properly checks the number of results.
bye, Sumit
Ack.
By the way, I think in general the sysdb API should behave consistently with respect to return codes of the basic operations. In other words, all search functions should return either ENOENT or EOK and msg->count == 0. Something similar would go for delete of an entry that doesn't exist.
I'd like to open a (minor) ticket to track this enhancement unless there much opposition.
On Tue, 2013-06-18 at 19:45 +0200, Jakub Hrozek wrote:
On Tue, Jun 18, 2013 at 05:30:09PM +0200, Sumit Bose wrote:
Hi,
one of the issues reported in https://bugzilla.redhat.com/show_bug.cgi?id=975001 is a core dump in the PAC responder. I thought that I can find missing results by checking for ENOENT which is wrong in this case. This patch properly checks the number of results.
bye, Sumit
Ack.
By the way, I think in general the sysdb API should behave consistently with respect to return codes of the basic operations. In other words, all search functions should return either ENOENT or EOK and msg->count == 0. Something similar would go for delete of an entry that doesn't exist.
I'd like to open a (minor) ticket to track this enhancement unless there much opposition.
please do, consistency is important.
Simo.
On Tue, Jun 18, 2013 at 04:15:54PM -0400, Simo Sorce wrote:
On Tue, 2013-06-18 at 19:45 +0200, Jakub Hrozek wrote:
On Tue, Jun 18, 2013 at 05:30:09PM +0200, Sumit Bose wrote:
Hi,
one of the issues reported in https://bugzilla.redhat.com/show_bug.cgi?id=975001 is a core dump in the PAC responder. I thought that I can find missing results by checking for ENOENT which is wrong in this case. This patch properly checks the number of results.
bye, Sumit
Ack.
By the way, I think in general the sysdb API should behave consistently with respect to return codes of the basic operations. In other words, all search functions should return either ENOENT or EOK and msg->count == 0. Something similar would go for delete of an entry that doesn't exist.
I'd like to open a (minor) ticket to track this enhancement unless there much opposition.
please do, consistency is important.
On Tue, Jun 18, 2013 at 07:45:51PM +0200, Jakub Hrozek wrote:
On Tue, Jun 18, 2013 at 05:30:09PM +0200, Sumit Bose wrote:
Hi,
one of the issues reported in https://bugzilla.redhat.com/show_bug.cgi?id=975001 is a core dump in the PAC responder. I thought that I can find missing results by checking for ENOENT which is wrong in this case. This patch properly checks the number of results.
bye, Sumit
Ack.
Pushed to master.
sssd-devel@lists.fedorahosted.org
-
Jakub Hrozek -
Simo Sorce -
Sumit Bose