8:04 p.m.
Get the values for the SystemName and SystemCreationClassName from the host instance and
set accordingly.
8:04 p.m.
New subject: [PATCH 1 of 2] Fix VSMigrationService NULL key attributes
# HG changeset patch
# User Kaitlin Rupert <karupert(a)us.ibm.com>
# Date 1197681878 28800
# Node ID 5339f01a786a287ed6f2c49b47704fd99db6432a
# Parent 72b93a4339e1c13080d6a6d80a580a7742c0b851
Fix VSMigrationService NULL key attributes.
The SystemCreationClassName and SystemName keys are NULL, these should be set.
Signed-off-by: Kaitlin Rupert <karupert(a)us.ibm.com>
diff -r 72b93a4339e1 -r 5339f01a786a src/Virt_VSMigrationService.c
--- a/src/Virt_VSMigrationService.c Fri Dec 14 10:37:01 2007 +0100
+++ b/src/Virt_VSMigrationService.c Fri Dec 14 17:24:38 2007 -0800
@@ -36,6 +36,7 @@
#include <libcmpiutil/std_invokemethod.h>
#include "Virt_VSMigrationService.h"
+#include "Virt_HostSystem.h"
#define CIM_JOBSTATE_STARTING 3
#define CIM_JOBSTATE_RUNNING 4
@@ -639,6 +640,12 @@ CMPIStatus get_migration_service(const C
{
CMPIInstance *inst;
CMPIStatus s = {CMPI_RC_OK, NULL};
+ CMPIInstance *host = NULL;
+ const char *val = NULL;
+
+ s = get_host_cs(broker, ref, &host);
+ if (s.rc != CMPI_RC_OK)
+ goto out;
inst = get_typed_instance(broker,
CLASSNAME(ref),
@@ -654,8 +661,30 @@ CMPIStatus get_migration_service(const C
CMSetProperty(inst, "Name",
(CMPIValue *)"MigrationService", CMPI_chars);
+ if (cu_get_str_prop(host, "Name", &val) != CMPI_RC_OK) {
+ cu_statusf(broker, &s,
+ CMPI_RC_ERR_FAILED,
+ "Unable to get name of HostSystem");
+ goto out;
+ }
+
+ CMSetProperty(inst, "SystemName",
+ (CMPIValue *)val, CMPI_chars);
+
+ if (cu_get_str_prop(host, "CreationClassName", &val) != CMPI_RC_OK)
{
+ cu_statusf(broker, &s,
+ CMPI_RC_ERR_FAILED,
+ "Unable to get creation class of HostSystem");
+ goto out;
+ }
+
+ CMSetProperty(inst, "SystemCreationClassName",
+ (CMPIValue *)val, CMPI_chars);
+
+
*_inst = inst;
+ out:
return s;
}
7:59 p.m.
New subject: [PATCH 1 of 2] Fix VSMigrationService NULL key attributes
Kaitlin Rupert wrote:
# HG changeset patch
# User Kaitlin Rupert <karupert(a)us.ibm.com>
# Date 1197681878 28800
# Node ID 5339f01a786a287ed6f2c49b47704fd99db6432a
# Parent 72b93a4339e1c13080d6a6d80a580a7742c0b851
Fix VSMigrationService NULL key attributes.
The SystemCreationClassName and SystemName keys are NULL, these should be set.
VirtualSystemMigrationSerive and VSMigrationService both need to get the
values for these attributes in the same way. It would be possible to
lump this bit of code into a function in Virt_HostSystem, but it's
unlikely that the implementation specific host provider will offer such
a function (since Virt_HostSystem is acting as a placeholder).
Thoughts?
--
Kaitlin Rupert
IBM Linux Technology Center
karupert(a)us.ibm.com
10:08 a.m.
New subject: [PATCH 1 of 2] Fix VSMigrationService NULL key attributes
KR> VirtualSystemMigrationSerive and VSMigrationService both need to
KR> get the values for these attributes in the same way. It would be
KR> possible to lump this bit of code into a function in
KR> Virt_HostSystem, but it's unlikely that the implementation
KR> specific host provider will offer such a function (since
KR> Virt_HostSystem is acting as a placeholder).
My plan was never to remove the placeholder, but rather to modify the
placeholder with logic to look up and return the proper host system.
This would give us a control point for switching between host
instrumentation, with the option to keep our fake one if nothing else
is available.
Adding such a function seems like a good idea to me, and as long as it
uses the existing internal logic in HostSystem to get the information,
it shouldn't need to change later.
Shall I hold off on this set or apply and let you add the generalized
function first?
Thanks!
--
Dan Smith
IBM Linux Technology Center
Open Hypervisor Team
email: danms(a)us.ibm.com
8:04 p.m.
New subject: [PATCH 2 of 2] Add libVirt_HostSystem to VSMigrationService build
# HG changeset patch
# User Kaitlin Rupert <karupert(a)us.ibm.com>
# Date 1197770616 28800
# Node ID b94dd9cf174968ecd58eeba26e37fc1a06dbdc4f
# Parent 5339f01a786a287ed6f2c49b47704fd99db6432a
Add libVirt_HostSystem to VSMigrationService build.
Signed-off-by: Kaitlin Rupert <karupert(a)us.ibm.com>
diff -r 5339f01a786a -r b94dd9cf1749 src/Makefile.am
--- a/src/Makefile.am Fri Dec 14 17:24:38 2007 -0800
+++ b/src/Makefile.am Sat Dec 15 18:03:36 2007 -0800
@@ -146,4 +146,6 @@ libVirt_ElementSettingData_la_LIBADD = -
libVirt_VSMigrationCapabilities_la_SOURCES = Virt_VSMigrationCapabilities.c
-libVirt_VSMigrationService_la_SOURCES = Virt_VSMigrationService.c
\ No newline at end of file
+libVirt_VSMigrationService_la_DEPENDENCIES = libVirt_HostSystem.la
+libVirt_VSMigrationService_la_SOURCES = Virt_VSMigrationService.c
+libVirt_VSMigrationService_la_LIBADD = -lVirt_HostSystem
3:24 a.m.
New subject: [PATCH 2 of 2] Add libVirt_HostSystem to VSMigrationService build
Kaitlin Rupert wrote:
# HG changeset patch
# User Kaitlin Rupert <karupert(a)us.ibm.com>
# Date 1197770616 28800
# Node ID b94dd9cf174968ecd58eeba26e37fc1a06dbdc4f
# Parent 5339f01a786a287ed6f2c49b47704fd99db6432a
Add libVirt_HostSystem to VSMigrationService build.
Signed-off-by: Kaitlin Rupert <karupert(a)us.ibm.com>
diff -r 5339f01a786a -r b94dd9cf1749 src/Makefile.am
--- a/src/Makefile.am Fri Dec 14 17:24:38 2007 -0800
+++ b/src/Makefile.am Sat Dec 15 18:03:36 2007 -0800
@@ -146,4 +146,6 @@ libVirt_ElementSettingData_la_LIBADD = -
libVirt_VSMigrationCapabilities_la_SOURCES = Virt_VSMigrationCapabilities.c
-libVirt_VSMigrationService_la_SOURCES = Virt_VSMigrationService.c
\ No newline at end of file
+libVirt_VSMigrationService_la_DEPENDENCIES = libVirt_HostSystem.la
+libVirt_VSMigrationService_la_SOURCES = Virt_VSMigrationService.c
+libVirt_VSMigrationService_la_LIBADD = -lVirt_HostSystem
Applying this patch crashes, if "[PATCH 3 of 3] AC no longer needs
libVirt_RASD" was applied before. I suppose the reason is that the "\ No
newline at end of file" was already fixed by the patch before.
--
Regards
Heidi Eckhart
Software Engineer
Linux Technology Center - Open Hypervisor
heidieck(a)linux.vnet.ibm.com
**************************************************
IBM Deutschland Entwicklung GmbH
Vorsitzender des Aufsichtsrats: Martin Jetter
Geschaeftsfuehrung: Herbert Kircher
Sitz der Gesellschaft: Boeblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294
3:25 a.m.
Kaitlin Rupert wrote:
Get the values for the SystemName and SystemCreationClassName from
the host instance and set accordingly.
Besides the minor problem with applying the changes to the makefile,
this works perfectly for me. Great work :). +1
--
Regards
Heidi Eckhart
Software Engineer
Linux Technology Center - Open Hypervisor
heidieck(a)linux.vnet.ibm.com
**************************************************
IBM Deutschland Entwicklung GmbH
Vorsitzender des Aufsichtsrats: Martin Jetter
Geschaeftsfuehrung: Herbert Kircher
Sitz der Gesellschaft: Boeblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294
6150
days inactive
6151
days old
6 comments
3 participants
participants (3)
-
Dan Smith -
Heidi Eckhart -
Kaitlin Rupert