[SSSD] [PATCH] Make it possible to inherit ignore_group_members, ldap_purge_cache_timeout and ldap_use_tokengroups into subdomains

Pavel Reichl preichl at redhat.com
Wed Jun 3 13:36:03 UTC 2015


Hello Jakub,

1st patch:
I remember Lukas having problems with such patches - to quote him:
> The most problematic part is that man page is updated before feature
> is implemented. Opposite order of patches could be acceptable
> but I prefer to have change in man page for small features in the same patch.
I'm CCing him if he wants to comment otherwise it's fine by me.
> +++ b/src/man/sssd.conf.5.xml
> @@ -479,7 +479,25 @@
>                          </para>
>                      </listitem>
>                  </varlistentry>
> -
> +                <varlistentry>
> +                    <term>subdomain_inherit (string)</term>
> +                    <listitem>
> +                        <para>
> +                            Specifies a list of configuration 
> parameters that
> +                            should be inherited by a subdomain. 
> Please note
> +                            that only selected parametrs can be 
> inherited.
typo:  paramet*e*rs
> +                        </para>
> +                        <para>

2nd patch)
Would you mind making parent_dom const?
> +void dp_option_inherit(struct sss_domain_info *parent_dom,
> +                       int option,
> +                       struct dp_option *parent_opts,
> +                       struct dp_option *subdom_opts)

3rd patch)
> +errno_t sdap_copy_map_entry(struct sdap_attr_map *src_map,
const?
> +                            struct sdap_attr_map *dst_map,
> +                            int entry_index)
> +{
> +    if (src_map[entry_index].name != NULL) {
> +        dst_map[entry_index].name = talloc_strdup(dst_map,
> + src_map[entry_index].name);
formatting issue
> +        if (dst_map[entry_index].name == NULL) {
> +            return ENOMEM;
> +        }
> +    } else {
> +        src_map->name = NULL;
                 dst_map?
> +    }
> +
> +    return EOK;
> +}
> +
I suppose that test_sdap_copy_map_entry_null_name needs dst_map not to 
be nulled to work properly.


5th patch)
> +static void sdap_inherit_user_options(struct sss_domain_info *parent_dom,
> +                                      struct sdap_attr_map 
> *parent_user_map,
> +                                      struct sdap_attr_map 
> *child_user_map)
> +{
> +    int inherit_options[] = {
> +        SDAP_AT_USER_PRINC,
> +        SDAP_OPTS_USER          /* sentinel */
> +    };
> +    int i;
> +
> +    for (i = 0; inherit_options[i] != SDAP_OPTS_USER; i++) {
> +        sdap_copy_map_entry(parent_user_map, child_user_map, i);
s/i/inherit_options[i] ?
> +    }
> +}

I'm going to test the patches now.

Thanks!



More information about the sssd-devel mailing list