From: "Mike Maslenkin" <mike.maslenkin@gmail.com>
To: Nickle Wang <nicklew@nvidia.com>
Cc: "devel@edk2.groups.io" <devel@edk2.groups.io>,
"abner.chang@amd.com" <abner.chang@amd.com>,
"igork@ami.com" <igork@ami.com>
Subject: Re: [edk2-devel] [edk2-redfish-client][PATCH 4/4] RedfishClientPkg: use POST method while provisioning new property.
Date: Mon, 18 Dec 2023 03:33:01 +0300 [thread overview]
Message-ID: <EE356099-04CD-4380-8B3D-6BEDFF51C54C@gmail.com> (raw)
In-Reply-To: <MW4PR12MB7031581D22691CD171305654D993A@MW4PR12MB7031.namprd12.prod.outlook.com>
[-- Attachment #1: Type: text/plain, Size: 5686 bytes --]
Hi Nickle,
> On 15. 12. 2023., at 04:53, Nickle Wang <nicklew@nvidia.com> wrote:
>
> Hi Mike,
>
> Per Redfish specification 7.9 POST(create)
>
> "The POST request is submitted to the resource collection to which the new resource will belong."
>
> If this is not a collection resource, we cannot use POST method. And /redfish/v1/Systems/SYS_ID/Bios is not a collection resource. The allowed method returned from BMC for BIOS resource is usually "GET" and "PUT".
>
> So, I think that the fourth parameter is still FALSE here. But I admit that the function header below is confusing and did not express above rule clearly.
>
> /**
> Provisioning redfish resource by given URI.
>
> @param[in] Schema Redfish schema information.
> @param[in] Uri Target URI to create resource.
> @param[in] InformationExchange Pointer to RESOURCE_INFORMATION_EXCHANGE.
> @param[in] HttpPostMode TRUE if resource does not exist, HTTP POST method is used.
> FALSE if the resource exist but some of properties are missing,
> HTTP PUT method is used.
>
> @retval EFI_SUCCESS Value is returned successfully.
> @retval Others Some error happened.
>
> **/
>
> Below is my suggestion.
>
> @param[in] HttpPostMode TRUE if target resource is a member of collection resource, HTTP POST method is used.
> FALSE if target resource is non-collection resource, HTTP PUT method is used.
>
> Do you think this helps to explain the use-case of fourth parameter more clearly?
>
> Thanks,
> Nickle
Seems like more comments need to be changed....
The idea behind this patch is the basis of the current implementation of RedfishClientPkg/Features/Bios/v1_0_9.
The EdkIIRedfishResourceConfigProvisioning function calls the EDKII_REDFISH_RESOURCE_CONFIG_PROTOCOL::Provisioning function,
this is RedfishResourceProvisioningResource() for this driver[1], then it calls RedfishProvisioningResourceCommon() where the logic
of this flag is completely changed [2].
So, for the current implementation of this flag, it is not a choice between POST and PUT, but between POST and PATCH (see [1]).
May be flag should not be inverted here [3]?
The value of FALSE here (you expect PUT to be used) means the following ProvisioningBiosExistResource() function calls
ProvisioningBiosProperties() with ProvisionMode == TRUE. This ProvisionMode == TRUE forces PropertyChanged set into TRUE,
so finally ProvisioningBiosProperties() returns success even for elements that do not exist.
Here I mean elements of /redfish/v1/Systems/{SystemID}/Bios/Attributes. I'm in situation when Attributes exists,
but it is empty.
Currently the PUT method is not used anywhere in RedfishClientPkg/Features/Bios and "PUT back to instance" actually performs the PATCH.
I will drop this patch from the current PR until it becomes clear how this can be improved.
Regards,
Mike.
>
>> -----Original Message-----
>> From: Mike Maslenkin <mike.maslenkin@gmail.com>
>> Sent: Friday, December 15, 2023 8:04 AM
>> To: devel@edk2.groups.io
>> Cc: abner.chang@amd.com; Nickle Wang <nicklew@nvidia.com>;
>> igork@ami.com; Mike Maslenkin <mike.maslenkin@gmail.com>
>> Subject: [edk2-redfish-client][PATCH 4/4] RedfishClientPkg: use POST method
>> while provisioning new property.
>>
>> External email: Use caution opening links or attachments
>>
>>
>> If EdkIIRedfishResourceConfigCheck fails according to the logic and
>> comment: new resources should be provisioned, so the POST method must be
>> used. Fourth parameter of EdkIIRedfishResourceConfigProvisioning is BOOLEAN
>> HttpPostMode, so we pass TRUE here.
>>
>> Cc: Abner Chang <abner.chang@amd.com>
>> Cc: Igor Kulchytskyy <igork@ami.com>
>> Cc: Nickle Wang <nicklew@nvidia.com>
>> Signed-off-by: Mike Maslenkin <mike.maslenkin@gmail.com>
>> ---
>> RedfishClientPkg/Features/Bios/v1_0_9/Common/BiosCommon.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/RedfishClientPkg/Features/Bios/v1_0_9/Common/BiosCommon.c
>> b/RedfishClientPkg/Features/Bios/v1_0_9/Common/BiosCommon.c
>> index a26a1083cd74..4fd4845f3420 100644
>> --- a/RedfishClientPkg/Features/Bios/v1_0_9/Common/BiosCommon.c
>> +++ b/RedfishClientPkg/Features/Bios/v1_0_9/Common/BiosCommon.c
>> @@ -818,9 +818,9 @@ HandleResource (
>> // The target property does not exist, do the provision to create property.
>>
>> //
>>
>> DEBUG ((REDFISH_DEBUG_TRACE, "%a provision for %s\n", __func__, Uri));
>>
>> - Status = EdkIIRedfishResourceConfigProvisioning (&SchemaInfo, Uri, Private-
>>> InformationExchange, FALSE);
>>
>> + Status = EdkIIRedfishResourceConfigProvisioning (&SchemaInfo, Uri,
>> + Private->InformationExchange, TRUE);
>>
>> if (EFI_ERROR (Status)) {
>>
>> - DEBUG ((DEBUG_ERROR, "%a, failed to provision with GET mode: %r\n",
>> __func__, Status));
>>
>> + DEBUG ((DEBUG_ERROR, "%a, failed to provision with POST mode:
>> + %r\n", __func__, Status));
>>
>> }
>>
>>
>>
>> return Status;
>>
>> --
>> 2.32.0 (Apple Git-132)
>
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#112610): https://edk2.groups.io/g/devel/message/112610
Mute This Topic: https://groups.io/mt/103181641/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-
[-- Attachment #2: Type: text/html, Size: 12041 bytes --]
next prev parent reply other threads:[~2023-12-18 0:33 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-15 0:03 [edk2-devel] [edk2-redfish-client][PATCH 0/4] change method for provisioning + minor fixes Mike Maslenkin
2023-12-15 0:03 ` [edk2-devel] [edk2-redfish-client][PATCH 1/4] RedfishClientPkg: add check for NULL pointer to avoid ASSERT Mike Maslenkin
2023-12-15 1:23 ` Nickle Wang via groups.io
2023-12-15 0:03 ` [edk2-devel] [edk2-redfish-client][PATCH 2/4] RedfishClientPkg: fix typo in EfiRestJsonStructureProtocolIsReady Mike Maslenkin
2023-12-15 1:23 ` Nickle Wang via groups.io
2023-12-15 0:03 ` [edk2-devel] [edk2-redfish-client][PATCH 3/4] RedfishClientPkg: fix access to unitialized variable Mike Maslenkin
2023-12-15 1:24 ` Nickle Wang via groups.io
2023-12-15 0:04 ` [edk2-devel] [edk2-redfish-client][PATCH 4/4] RedfishClientPkg: use POST method while provisioning new property Mike Maslenkin
2023-12-15 1:53 ` Nickle Wang via groups.io
2023-12-18 0:33 ` Mike Maslenkin [this message]
[not found] ` <17A1C5C572E0EF8A.24236@groups.io>
2023-12-18 0:57 ` Mike Maslenkin
2023-12-18 5:32 ` Chang, Abner via groups.io
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=EE356099-04CD-4380-8B3D-6BEDFF51C54C@gmail.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox