[libvirt] [PATCH] ESX: append CURL headers to fix serviceContent entities

[Matthias Bolte]

2012/12/31 Ata Bohra <ata.husain at hotmail.com>:
> Please hold off reviews on this post. I've been testing and analyzing some
> other piece of code (the one that do not need connection headers to include
> urn:vim25 namespace), but I observed that this namespace include causes lot
> of other issues such as:
> 1. It breaks existing APIs due to ordering issues of Object variables during
> serialization.

Yes, the ESX server is picky about the ordering here. That's a problem.

> 2. Introduces/exposes lot of other fields that were not present earlier
> causing "warning" messages during deserialization (this one is benign, but
> first one is FATAL).

The warnings about unexpected fields are leftovers from the early
development stage of the ESX driver. They can be removed if the get in
the way.

> BUT, if this namespace is not included it not only prevents populating of
> ServiceContent object but also hides some API methods such as:
> ParseDescriptor, ImportVApp and CreateImportSpec etc. I'm thinking of
> limiting the impact and update the headers only inside the routines where
> its needed and revert back the headers on exit (given it is necessary evil).
> Please suggest if there is a better idea.

Currently the ESX driver uses only functions and objects from the
vSphere API 2.5 (at least it should). This allows it to deal with
older ESX server versions down to 3.5. What you're trying to do now is
using functions added in a newer vSphere API, don't you? I assume that
all the fields and functions you're talking about, that require the
SOAPAction header, were added in vSphere API 4.0 or later.

Is it documented somewhere that you need to pass the SOAPAction header
to the server in order to access vSphere API > 2.5 features?

But the real problem could be the serialization order. If this is
really different between vSphere API 2.5 and 4.x than we need to
decide this at runtime and also extend the generator to know bot
orderings. This can become very ugly.


> Thanks !
> Ata
>
>> From: ata.husain at hotmail.com
>> To: libvir-list at redhat.com
>> CC: ata.husain at hotmail.com
>> Subject: [PATCH] ESX: append CURL headers to fix serviceContent entities
>> Date: Tue, 25 Dec 2012 16:44:23 -0800
>
>>
>> Append curl headers with "SOAPAction" header to populate
>> serviceContent object entities which are otherwise missing.
>> ---
>>  src/esx/esx_vi.c               | 2 ++
>>  src/esx/esx_vi_generator.input | 3 +++
>> 2 files changed, 5 insertions(+)
>>
>> diff --git a/src/esx/esx_vi.c b/src/esx/esx_vi.c
>> index 99c1eb1..a379183 100644
>> --- a/src/esx/esx_vi.c
>> +++ b/src/esx/esx_vi.c
>> @@ -322,6 +322,8 @@ esxVI_CURL_Connect(esxVI_CURL *curl, esxUtil_ParsedUri
>> *parsedUri)
>> * approx. 2 sec per POST operation.
>> */
>> curl->headers = curl_slist_append(curl->headers, "Expect:");
>> + curl->headers = curl_slist_append(curl->headers,
>> + _("SOAPAction: \"urn:vim25\""));
>>
>> if (curl->headers == NULL) {
>>          virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
>> diff --git a/src/esx/esx_vi_generator.input
>> b/src/esx/esx_vi_generator.input
>> index 22c114e..236dcb3 100644
>> --- a/src/esx/esx_vi_generator.input
>> +++ b/src/esx/esx_vi_generator.input
>> @@ -1008,6 +1008,8 @@ object ServiceContent
>>      ManagedObjectReference                   clusterProfileManager
>> o
>>      ManagedObjectReference                   complianceManager
>> o
>>      ManagedObjectReference                   localizationManager
>> o
>> +    ManagedObjectReference                   storageResourceManager
>> o
>> +    ManagedObjectReference                   guestOperationsManager
>> o
>> end
>>
>>
>> @@ -1073,6 +1075,7 @@ object UserSession
>>      DateTime                                 lastActiveTime
>> r
>> String locale                         r
>> String                                   messageLocale                  r
>> + Boolean                                  extensionSession
>> o
>> end
>>
>>
>> --
>> 1.7.9.5
>>
>
> --
> libvir-list mailing list
> libvir-list at redhat.com
> https://www.redhat.com/mailman/listinfo/libvir-list



-- 
Matthias Bolte
http://photron.blogspot.com