Also add list_mandatory_options() to both SSSDService and SSSDDomain objects.
There is a new list_options_with_mandatory() function that will return a longer tuple than list_options(), including the mandatory flag directly.
list_options() will still return the shorter tuple, for backwards-compatibility.
This feature will allow authconfig (a consumer of the SSSDConfig API) to produce a simpler interface listing only the mandatory options for each provider. This should result in a less confusing GUI.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 01/04/2010 07:31 PM, Stephen Gallagher wrote:
Also add list_mandatory_options() to both SSSDService and SSSDDomain objects.
There is a new list_options_with_mandatory() function that will return a longer tuple than list_options(), including the mandatory flag directly.
list_options() will still return the shorter tuple, for backwards-compatibility.
This feature will allow authconfig (a consumer of the SSSDConfig API) to produce a simpler interface listing only the mandatory options for each provider. This should result in a less confusing GUI.
Ack,
but as discussed off-list it might be nice to amend the commit message with explanation on why was the default value for debug_level and reconnection_retries in [services] removed with the patch.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 01/05/2010 11:11 AM, Jakub Hrozek wrote:
On 01/04/2010 07:31 PM, Stephen Gallagher wrote:
Also add list_mandatory_options() to both SSSDService and SSSDDomain objects.
There is a new list_options_with_mandatory() function that will return a longer tuple than list_options(), including the mandatory flag directly.
list_options() will still return the shorter tuple, for backwards-compatibility.
This feature will allow authconfig (a consumer of the SSSDConfig API) to produce a simpler interface listing only the mandatory options for each provider. This should result in a less confusing GUI.
Ack,
but as discussed off-list it might be nice to amend the commit message with explanation on why was the default value for debug_level and reconnection_retries in [services] removed with the patch.
Address https://fedorahosted.org/sssd/ticket/349
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded. Patch 0003: Implement the mandatory flag as in the previous patch.
- -- Stephen Gallagher RHCE 804006346421761
Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 01/07/2010 05:54 PM, Stephen Gallagher wrote:
Address https://fedorahosted.org/sssd/ticket/349
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded.
Ack
Patch 0003: Implement the mandatory flag as in the previous patch.
Nack,
the schema adds ipa_hostname as mandatory service.
Also all of the {access,auth,chpass,id}_providers are marked as mandatory - is that correct? A domain with just id and auth providers is a perfectly valid one.
In the LDAP config, I think plain ldap_search_base should be marked as mandatory, not the specialised ldap_{user,group}_search base - if these are not set, the value of ldap_search_base is used.
IIRC the krb5_realm and krb5_kdcip is not strictly mandatory when using the locator plugin, but I guess that the recommended use case would be to put the info into our config, so these are probably fine.
The code itself looks fine to me.
Jakub
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 01/11/2010 09:58 AM, Jakub Hrozek wrote:
On 01/07/2010 05:54 PM, Stephen Gallagher wrote:
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded.
Ack
Patch 0003: Implement the mandatory flag as in the previous patch.
Nack,
the schema adds ipa_hostname as mandatory service.
Also all of the {access,auth,chpass,id}_providers are marked as mandatory - is that correct? A domain with just id and auth providers is a perfectly valid one.
In the LDAP config, I think plain ldap_search_base should be marked as mandatory, not the specialised ldap_{user,group}_search base - if these are not set, the value of ldap_search_base is used.
IIRC the krb5_realm and krb5_kdcip is not strictly mandatory when using the locator plugin, but I guess that the recommended use case would be to put the info into our config, so these are probably fine.
The code itself looks fine to me.
Jakub
I think I'm going to withdraw these patches for now. I think we probably need more fine-grained control on these options, such as "Mandatory, optional and advanced" (where ipa_domain might be optional, depending on whether or not the system is using DNS autodiscovery when we support it).
- -- Stephen Gallagher RHCE 804006346421761
Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 01/11/2010 09:58 AM, Jakub Hrozek wrote:
On 01/07/2010 05:54 PM, Stephen Gallagher wrote:
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded.
Ack
Patch 0003: Implement the mandatory flag as in the previous patch.
Nack,
the schema adds ipa_hostname as mandatory service.
Also all of the {access,auth,chpass,id}_providers are marked as mandatory - is that correct? A domain with just id and auth providers is a perfectly valid one.
I made chpass_provider and access_provider optional
In the LDAP config, I think plain ldap_search_base should be marked as mandatory, not the specialised ldap_{user,group}_search base - if these are not set, the value of ldap_search_base is used.
I reversed these, so now ldap_search_base is the only mandatory option.
IIRC the krb5_realm and krb5_kdcip is not strictly mandatory when using the locator plugin, but I guess that the recommended use case would be to put the info into our config, so these are probably fine.
Left them alone.
The code itself looks fine to me.
Jakub
Per the most recent discussions on authconfig integration, we still need this flag.
- -- Stephen Gallagher RHCE 804006346421761
Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 02/04/2010 09:20 PM, Stephen Gallagher wrote:
On 01/11/2010 09:58 AM, Jakub Hrozek wrote:
On 01/07/2010 05:54 PM, Stephen Gallagher wrote:
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded.
Ack
This patch is the same, so the ACK remains.
Patch 0003: Implement the mandatory flag as in the previous patch.
Nack,
the schema adds ipa_hostname as mandatory service.
Also all of the {access,auth,chpass,id}_providers are marked as mandatory - is that correct? A domain with just id and auth providers is a perfectly valid one.
I made chpass_provider and access_provider optional
The only glitch I found was that in order for the unittests to keep working, you also need to remove chpass_provider and access_provider from SSSDConfigTestSSSDDomain.testListMandatoryOptions()
The other changes look fine to me, and FTR, they are the only changes versus the previous patch. The code looks fine to me.
Jakub
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 02/05/2010 08:12 AM, Jakub Hrozek wrote:
I made chpass_provider and access_provider optional
The only glitch I found was that in order for the unittests to keep working, you also need to remove chpass_provider and access_provider from SSSDConfigTestSSSDDomain.testListMandatoryOptions()
The other changes look fine to me, and FTR, they are the only changes versus the previous patch. The code looks fine to me.
That's what I get for not rerunning the test suite...
New patch attached.
- -- Stephen Gallagher RHCE 804006346421761
Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 02/05/2010 02:18 PM, Stephen Gallagher wrote:
New patch attached.
ACK
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 02/05/2010 08:49 AM, Jakub Hrozek wrote:
On 02/05/2010 02:18 PM, Stephen Gallagher wrote:
New patch attached.
ACK
Pushed to master.
- -- Stephen Gallagher RHCE 804006346421761
Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/
On Thu, 2010-01-07 at 11:54 -0500, Stephen Gallagher wrote:
Address https://fedorahosted.org/sssd/ticket/349
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded. Patch 0003: Implement the mandatory flag as in the previous patch.
Is the mandatory flag really the way how to proceed with the authconfig?
Will there be a short summary of the Friday authconfig SSSD UI meeting posted somewhere?
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 01/11/2010 10:07 AM, Tomas Mraz wrote:
On Thu, 2010-01-07 at 11:54 -0500, Stephen Gallagher wrote:
Address https://fedorahosted.org/sssd/ticket/349
After a bit of thought, I decided to break this into two separate patches to make it more clear. New patches attached.
Patch 0002: Remove the explicit defaults where they're unneeded. Patch 0003: Implement the mandatory flag as in the previous patch.
Is the mandatory flag really the way how to proceed with the authconfig?
Will there be a short summary of the Friday authconfig SSSD UI meeting posted somewhere?
Yeah, as I mentioned in my other reply, I think this is no longer the correct approach.
I'm withdrawing this and I want to work from the results of that meeting. I will ping the other attendants about the notes.
- -- Stephen Gallagher RHCE 804006346421761
Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/
sssd-devel@lists.fedorahosted.org
-
Jakub Hrozek -
Stephen Gallagher -
Stephen Gallagher -
Tomas Mraz