sblim-issues — Mailing list for bugtracker activity

[Sblim-issues] Bug - 2760: wbemcli query truncates the password if it has '@' character

[Ashok VI <ash...@gm...>]

Hi Team,

I have filed below issue. Can someone confirm if it indeed a bug or not?

https://sourceforge.net/p/sblim/bugs/2760/

Thanks,
Ashok

[Sblim-issues] (no subject)

[ajay g. <aja...@gm...>]

Hi,

Can anyone help me to understand these two issues-

1.   3001896 "httpAdapter can crash on bad content length"
2.   3001915 "Possible heap corruption in httpAdapter"

both are resolved in sblim-sfcb-1.3.8.
Actually I wanted to reproduce the bug in sblim-sfcb-1.2.4
So any suggestion on how to reproduce the above issues?

Thanks & Regards,
Ajay Gupta

[Sblim-issues] sblim-cmpi-sysfs: Cannot get scsi_device or scsi_host instances

[Stewart, S. <Sea...@ne...>]

Hi all,

While playing around with the Linux_SysfsSCSIDevice and Linux_SysfsSCSIHostDevice classes on SLES12,
I noticed that I would get no instances back when trying to enumerate their instance names.  After a
little investigation I found that this code was the problem:

void * Linux_SysfsDeviceUtil_beginEnumeration( void * location )
{
   FILE * handle;
   char * findcommand = malloc(strlen((char *)location)+38);

   /* Find all the instance names in the target location */
   sprintf(findcommand, "find %s -maxdepth 1 -mindepth 1 -type d", (char *)location);
   handle = popen(findcommand,"r");
   free(findcommand);
   return handle;
}

The find command needs -L in order to get the symlinks (all the scsi_device and scsi_host instances
are symlinks).

I searched around and found this list, but am unsure of processes for submitting bug fixes to any
of these projects.  I see that version 1.20 is still the latest for the sysfs provider, so this
problem still exists.  Are these packages still maintained?  I might also be interested in contributing
a couple more providers in the future.

I would appreciate any nudge in the right direction that anyone could provide. 


Thanks!

Sean Stewart
Linux Software Engineer
Hyperscale Storage Group
NetApp

[Sblim-issues] create indication subscription on local sfcb fails (using sfcc)

[jianbo p. <rus...@gm...>]

Hi,
I create filter and handler successfully on sfcb by using sfcc. But it
fails when trying creating subscription. CreateInstance() returns 7, which
means that "No supported indication classes in filter query or no provider
found". Anyone has any idea why this happens?
Below is my code. Many thanks.
void SfcbClient::createFilter()
{
        logger.info("enter createFilter");
        CMPIObjectPath * objectpath = NULL, * objectpath2 = NULL;
        CMPIInstance * instance = NULL;
        CMPIStatus status;

        objectpath = newCMPIObjectPath("root/interop",
"CIM_IndicationFilter", NULL);
        CMAddKey(objectpath, "CreationClassName", "CIM_IndicationFilter",
CMPI_chars);
        CMAddKey(objectpath, "SystemName", "localhost", CMPI_chars);
        CMAddKey(objectpath, "SystemCreationClassName",
"CIM_ComputerSystem", CMPI_chars);
        CMAddKey(objectpath, "Name", "hpshd", CMPI_chars);
        CMAddKey(objectpath, "SourceNamespace", "root/interop", CMPI_chars);
        //CMAddKey(objectpath, "Query", "SELECT * FROM
CIM_ProcessIndication", CMPI_chars);
        CMAddKey(objectpath, "Query", "SELECT * FROM CIM_OperatingSystem",
CMPI_chars);
        CMAddKey(objectpath, "QueryLanguage", "WQL", CMPI_chars);

        instance = newCMPIInstance(objectpath, NULL);
        //CMSetProperty(instance, "Name", "hpshd", CMPI_chars);
        //CMSetProperty(instance, "Query", "SELECT * FROM
CIM_ProcessIndication", CMPI_chars);
        //CMSetProperty(instance, "QueryLanguage", "WQL", CMPI_chars);

        objectpath2 = pclient->ft->createInstance(pclient, objectpath,
instance, &status);
        filterPath = objectpath2;

        logger.info("createInstance() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);
        if(status.rc == 11)
        {
                status = pclient->ft->deleteInstance(pclient, objectpath);
                logger.info("deleteInstance() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);
        }
        //if (instance) CMRelease(instance);
        //if (objectpath) CMRelease(objectpath);
        //if (status.msg) CMRelease(status.msg);
        logger.info("exit createFilter");
}
void SfcbClient::createHandler()
{
        logger.info("enter createHandler");
        CMPIObjectPath * objectpath = NULL, * objectpath2 = NULL;
        CMPIInstance * instance = NULL;
        CMPIStatus status;

        objectpath = newCMPIObjectPath("root/interop",
"CIM_IndicationHandlerCIMXML", NULL);
        CMAddKey(objectpath, "CreationClassName",
"CIM_IndicationHandlerCIMXML", CMPI_chars);
        CMAddKey(objectpath, "SystemName", "localhost", CMPI_chars);
        CMAddKey(objectpath, "SystemCreationClassName",
"CIM_ComputerSystem", CMPI_chars);
        CMAddKey(objectpath, "Name", "hpshd", CMPI_chars);

        instance = newCMPIInstance(objectpath, NULL);
        CMSetProperty(instance, "Name", "hpshd", CMPI_chars);
        stringstream ss;
        ss << "http://localhost:" << portNumber;
        CMSetProperty(instance, "Destination", ss.str().c_str(),
CMPI_chars);

        objectpath2 = pclient->ft->createInstance(pclient, objectpath,
instance, &status);
        handlerPath = objectpath2;

        logger.info("createInstance() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);
        if(status.rc == 11)
        {
                status = pclient->ft->deleteInstance(pclient, objectpath);
                logger.info("deleteInstance() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);
        }
        //if (instance) CMRelease(instance);
        //if (objectpath) CMRelease(objectpath);
        //if (status.msg) CMRelease(status.msg);
        logger.info("exit createHandler");
}

void SfcbClient::createSubscription()
{
        createFilter();
        createHandler();
        logger.info("enter createSubscription");
        CMPIObjectPath * objectpath, * objectpath2;
        CMPIInstance * instance;
        CMPIStatus status;

        objectpath = newCMPIObjectPath("root/interop",
"CIM_IndicationSubscription", &status);
        CMPIValue value;
        value.ref = filterPath;
        CMAddKey(objectpath, "Filter", &value, CMPI_ref);

        value.ref = handlerPath;
        CMAddKey(objectpath, "Handler", &value, CMPI_ref);
        //CMAddKey(objectpath, "Filter", filterPath, CMPI_ref);
        //CMAddKey(objectpath, "Handler", handlerPath, CMPI_ref);

        logger.info("newCMPIObjectPath() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);
        instance = newCMPIInstance(objectpath, &status);
#if 1
        value.uint16 = 2;
        CMSetProperty(instance, "OnFatalErrorPolicy", &value, CMPI_uint16);

        value.uint16 = 2;
        CMSetProperty(instance, "SubscriptionState", &value, CMPI_uint16);

        value.uint16 = 2;
        CMSetProperty(instance, "RepeatNotificationPolicy", &value,
CMPI_uint16);
#endif
#if 0
        value.ref = filterPath;
        CMSetProperty(instance, "Filter", &value, CMPI_ref);
        value.ref = handlerPath;
      CMSetProperty(instance, "Handler", &value, CMPI_ref);
#endif
//        CMSetProperty(instance, "Filter", filterPath, CMPI_ref);
 //     CMSetProperty(instance, "Handler", handlerPath, CMPI_ref);
        logger.info("newCMPIInstance() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);

        objectpath2 = pclient->ft->createInstance(pclient, objectpath,
instance, &status);
        CMRelease(objectpath);
        objectpath = objectpath2;

        logger.info("createInstance() rc=%d, msg=%s\n", status.rc,
(status.msg)? (char *)status.msg->hdl : NULL);
        if (instance) CMRelease(instance);
        if (objectpath) CMRelease(objectpath);
        if (status.msg) CMRelease(status.msg);
        logger.info("exit createSubscription");

}

[Sblim-issues] [sblim:sfcb-tix] Mass edit changes by Chris Buccella

[Chris B. <buc...@us...>]

Mass edit changing:

- **Release**: 1.4.6


ticket: sfcb-tix:#9 gatherNameSpacesData and gatherNameSpaces similar

- **Release**: 1.4.5 --> 1.4.6


ticket: sfcb-tix:#20 SFCB does not honor LocalOnly for class operations

- **Release**: 1.4.5 --> 1.4.6


ticket: sfcb-tix:#23 Pitfalls in usage of _SFCB_ABORT()

- **Release**: 1.4.5 --> 1.4.6


ticket: sfcb-tix:#10 local interface memory leak in getobjectpath

- **Release**: 1.4.5 --> 1.4.6


ticket: sfcb-tix:#57 sblim-cmpi-smbios package availability for RHEL 6.2 or 6.3

- **Release**: 1.4.5 --> 1.4.6


ticket: sfcb-tix:#22 Alternatives to using a hard-coded pathname for ftok()

- **Release**: 1.4.5 --> 1.4.6

[Sblim-issues] [sblim:sfcb-tix] Mass edit changes by Dave Heller

[Dave H. <hel...@us...>]

Mass edit changing:

- **Status**: fixed


ticket: sfcb-tix:#11 Remove unnecessary casts for allocs

- **Status**: pending --> fixed


ticket: sfcb-tix:#36 Add Pulled Enums test provider

- **Status**: pending --> fixed


ticket: sfcb-tix:#46 Error parsing CDATA with no whitespace

- **Status**: pending --> fixed


ticket: sfcb-tix:#49 Cannot return EmbeddedInstance property from provider

- **Status**: pending --> fixed


ticket: sfcb-tix:#43 Unnecessary EI during interopProvider init

- **Status**: pending --> fixed


ticket: sfcb-tix:#41 Global IndicationService

- **Status**: pending --> fixed


ticket: sfcb-tix:#40 Add Valgrind test

- **Status**: pending --> fixed


ticket: sfcb-tix:#31 SFCB TestAssociationProvider can cause some queries to fail

- **Status**: pending --> fixed


ticket: sfcb-tix:#47 Pulled Enumeration Support - CIMXML Parser

- **Status**: pending --> fixed


ticket: sfcb-tix:#32 Provider Mgr trace should identify which provider is being called

- **Status**: pending --> fixed


ticket: sfcb-tix:#17 mofc doesn't unescape characters

- **Status**: pending --> fixed


ticket: sfcb-tix:#42 Make sure setupControl has run prior to getControl

- **Status**: pending --> fixed


ticket: sfcb-tix:#2 Problem with PropertyList on association classes.

- **Status**: pending --> fixed


ticket: sfcb-tix:#12 Add http adapter support for a Expect: 100-continue header.

- **Status**: pending --> fixed


ticket: sfcb-tix:#18 disable default properties breaks test

- **Status**: pending --> fixed


ticket: sfcb-tix:#48 mofc: Allow absoltute paths in #pragma include

- **Status**: pending --> fixed

[Sblim-issues] [ sblim-Bugs-3529678 ] sfcc getClass does not expose method information

[SourceForge.net <no...@so...>]

Bugs item #3529678, was opened at 2012-05-25 00:39
Message generated for change (Settings changed) made by hellerda
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcc
Group: New Feature
>Status: Pending
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: Klaus Kämpf (kkaempf)
Assigned to: Dave Heller (hellerda)
Summary: sfcc getClass does not expose method information

Initial Comment:
Retrieving class information with getClass only exposes property and qualifier information but not method information.


----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-07 17:30

Message:
Pushed to git master

----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-05 16:24

Message:
Patch for final review.  Adds support for class methods, including method
parameters and qualifiers.  Support for parameter references, qualifier
value arrays, deferred as future TODO.

----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-04 09:04

Message:
Preliminary patch attached.  This adds support for class methods, including
method parameters and qualifiers.  The program "TEST/test_gc" is expanded
to show the results.  TODO:
- Handle parameter references (currently they are parsed but not handled in
the parameter modeling)
- Handle qualifier value arrays (currently handles only values, not value
arrays)
- Add new error condition: CMPI_RC_ERR_NO_SUCH_METHOD
- Expand (and relocate?) utility function type2Chars()


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

[Sblim-issues] [ sblim-Bugs-3529678 ] sfcc getClass does not expose method information

[SourceForge.net <no...@so...>]

Bugs item #3529678, was opened at 2012-05-25 00:39
Message generated for change (Comment added) made by hellerda
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcc
Group: New Feature
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Klaus Kämpf (kkaempf)
Assigned to: Dave Heller (hellerda)
Summary: sfcc getClass does not expose method information

Initial Comment:
Retrieving class information with getClass only exposes property and qualifier information but not method information.


----------------------------------------------------------------------

>Comment By: Dave Heller (hellerda)
Date: 2013-02-07 17:30

Message:
Pushed to git master

----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-05 16:24

Message:
Patch for final review.  Adds support for class methods, including method
parameters and qualifiers.  Support for parameter references, qualifier
value arrays, deferred as future TODO.

----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-04 09:04

Message:
Preliminary patch attached.  This adds support for class methods, including
method parameters and qualifiers.  The program "TEST/test_gc" is expanded
to show the results.  TODO:
- Handle parameter references (currently they are parsed but not handled in
the parameter modeling)
- Handle qualifier value arrays (currently handles only values, not value
arrays)
- Add new error condition: CMPI_RC_ERR_NO_SUCH_METHOD
- Expand (and relocate?) utility function type2Chars()


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

[Sblim-issues] [ sblim-Bugs-3196679 ] gather: remove sysfs dependency when not building z plugins

[SourceForge.net <no...@so...>]

Bugs item #3196679, was opened at 2011-03-01 15:51
Message generated for change (Comment added) made by tyreld
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3196679&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Metric Gatherer
Group: Usability
>Status: Pending
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: Tyrel Datwyler (tyreld)
Assigned to: Tyrel Datwyler (tyreld)
Summary: gather: remove sysfs dependency when not building z plugins

Initial Comment:
The dependency on libsysfs only exists when building plugins for System z metrics. Modern distros ship libsysfs, but removing the dependency allows the gatherer to run on older x86 kernels prior to sysfs.

----------------------------------------------------------------------

>Comment By: Tyrel Datwyler (tyreld)
Date: 2013-02-07 16:32

Message:
Pushed to git master

----------------------------------------------------------------------

Comment By: Tyrel Datwyler (tyreld)
Date: 2013-02-07 16:26

Message:
Attached patch.

Dependent on fix from tracker item 3603719

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3196679&group_id=128809

[Sblim-issues] [ sblim-Bugs-3196679 ] gather: remove sysfs dependency when not building z plugins

[SourceForge.net <no...@so...>]

Bugs item #3196679, was opened at 2011-03-01 15:51
Message generated for change (Comment added) made by tyreld
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3196679&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Metric Gatherer
Group: Usability
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Tyrel Datwyler (tyreld)
Assigned to: Tyrel Datwyler (tyreld)
Summary: gather: remove sysfs dependency when not building z plugins

Initial Comment:
The dependency on libsysfs only exists when building plugins for System z metrics. Modern distros ship libsysfs, but removing the dependency allows the gatherer to run on older x86 kernels prior to sysfs.

----------------------------------------------------------------------

>Comment By: Tyrel Datwyler (tyreld)
Date: 2013-02-07 16:26

Message:
Attached patch.

Dependent on fix from tracker item 3603719

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3196679&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603719 ] Clean up configuration toolchain and fix DSO linking

[SourceForge.net <no...@so...>]

Bugs item #3603719, was opened at 2013-02-07 07:40
Message generated for change (Comment added) made by tyreld
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603719&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Metric Gatherer
>Group: Code Cleanup
>Status: Pending
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: Michael Chase-Salerno (mchasal)
Assigned to: Tyrel Datwyler (tyreld)
Summary: Clean up configuration toolchain and fix DSO linking

Initial Comment:
On the latest Ubuntu 12.10, I get the following errors during build:

libtool: link: gcc -Wall -g -O2 -o .libs/gatherctl gatherctl.o  ./.libs/librgather.so -lvirt -lsysfs -ldl -lpthread
/usr/bin/ld: gatherctl.o: undefined reference to symbol 'ch_release'
/usr/bin/ld: note: 'ch_release' is defined in DSO /home/msalerno/CODE/gather/.libs/libgatherutil.so.0 so try adding it to the linker command line
/home/msalerno/CODE/gather/.libs/libgatherutil.so.0: could not read symbols: Invalid operation
collect2: error: ld returned 1 exit status

I've run into this with other sblim projects as well and added some addition dependencies to address it, the attached patch cleans them up, but I understand that there may be a different approach.

----------------------------------------------------------------------

>Comment By: Tyrel Datwyler (tyreld)
Date: 2013-02-07 15:49

Message:
Pushed to git master

----------------------------------------------------------------------

Comment By: Tyrel Datwyler (tyreld)
Date: 2013-02-07 14:57

Message:
The following discusses the toolchain change that removed implicit DSO
linking in the distros:

http://fedoraproject.org/wiki/Features/ChangeInImplicitDSOLinking

I'm attaching an updated patch the contains the already attached DSO fix as
well as some clean up of the autotools configuration files.

1.) replace deprecated AC_PROG_LIBTOOL with LT_INIT
2.) fix over linking by setting component specific lib variables in
AC_CHECK_LIB instead of allowing the default action of appending found
libraries to global LIBS
3.) remove redundant _DEPENDENCIES rules since any library listed in
_LIBADD is automatically added to the dependency list

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603719&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603719 ] Clean up configuration toolchain and fix DSO linking

[SourceForge.net <no...@so...>]

Bugs item #3603719, was opened at 2013-02-07 07:40
Message generated for change (Comment added) made by tyreld
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603719&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Metric Gatherer
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Michael Chase-Salerno (mchasal)
Assigned to: Tyrel Datwyler (tyreld)
>Summary: Clean up configuration toolchain and fix DSO linking

Initial Comment:
On the latest Ubuntu 12.10, I get the following errors during build:

libtool: link: gcc -Wall -g -O2 -o .libs/gatherctl gatherctl.o  ./.libs/librgather.so -lvirt -lsysfs -ldl -lpthread
/usr/bin/ld: gatherctl.o: undefined reference to symbol 'ch_release'
/usr/bin/ld: note: 'ch_release' is defined in DSO /home/msalerno/CODE/gather/.libs/libgatherutil.so.0 so try adding it to the linker command line
/home/msalerno/CODE/gather/.libs/libgatherutil.so.0: could not read symbols: Invalid operation
collect2: error: ld returned 1 exit status

I've run into this with other sblim projects as well and added some addition dependencies to address it, the attached patch cleans them up, but I understand that there may be a different approach.

----------------------------------------------------------------------

>Comment By: Tyrel Datwyler (tyreld)
Date: 2013-02-07 14:57

Message:
The following discusses the toolchain change that removed implicit DSO
linking in the distros:

http://fedoraproject.org/wiki/Features/ChangeInImplicitDSOLinking

I'm attaching an updated patch the contains the already attached DSO fix as
well as some clean up of the autotools configuration files.

1.) replace deprecated AC_PROG_LIBTOOL with LT_INIT
2.) fix over linking by setting component specific lib variables in
AC_CHECK_LIB instead of allowing the default action of appending found
libraries to global LIBS
3.) remove redundant _DEPENDENCIES rules since any library listed in
_LIBADD is automatically added to the dependency list

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603719&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603454 ] control should do type conversions on startup

[SourceForge.net <no...@so...>]

Bugs item #3603454, was opened at 2013-02-05 12:31
Message generated for change (Settings changed) made by buccella
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcb
Group: Performance
>Status: Pending
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Chris Buccella (buccella)
Assigned to: Chris Buccella (buccella)
Summary: control should do type conversions on startup

Initial Comment:
The control values (from sfcb.cfg) are stored in a hashtable, with the values being strings. Calls to getControlNum() will perform a strtol() call on the string value every time it is called. Similarly, getControlBool() will perform a strcmp() every time it is called. This is inefficient, since we never need to use those values as strings. Instead, do the conversions once (at startup) and store the useful data types on the hashtable.

----------------------------------------------------------------------

Comment By: Chris Buccella (buccella)
Date: 2013-02-07 12:51

Message:
NOT committed to HEAD, but pushed to git master

----------------------------------------------------------------------

Comment By: Chris Buccella (buccella)
Date: 2013-02-07 12:50

Message:
committed to HEAD

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603454 ] control should do type conversions on startup

[SourceForge.net <no...@so...>]

Bugs item #3603454, was opened at 2013-02-05 12:31
Message generated for change (Settings changed) made by buccella
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcb
Group: Performance
>Status: Open
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Chris Buccella (buccella)
Assigned to: Chris Buccella (buccella)
Summary: control should do type conversions on startup

Initial Comment:
The control values (from sfcb.cfg) are stored in a hashtable, with the values being strings. Calls to getControlNum() will perform a strtol() call on the string value every time it is called. Similarly, getControlBool() will perform a strcmp() every time it is called. This is inefficient, since we never need to use those values as strings. Instead, do the conversions once (at startup) and store the useful data types on the hashtable.

----------------------------------------------------------------------

>Comment By: Chris Buccella (buccella)
Date: 2013-02-07 12:51

Message:
NOT committed to HEAD, but pushed to git master

----------------------------------------------------------------------

Comment By: Chris Buccella (buccella)
Date: 2013-02-07 12:50

Message:
committed to HEAD

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603454 ] control should do type conversions on startup

[SourceForge.net <no...@so...>]

Bugs item #3603454, was opened at 2013-02-05 12:31
Message generated for change (Comment added) made by buccella
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcb
Group: Performance
>Status: Pending
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: Chris Buccella (buccella)
Assigned to: Chris Buccella (buccella)
Summary: control should do type conversions on startup

Initial Comment:
The control values (from sfcb.cfg) are stored in a hashtable, with the values being strings. Calls to getControlNum() will perform a strtol() call on the string value every time it is called. Similarly, getControlBool() will perform a strcmp() every time it is called. This is inefficient, since we never need to use those values as strings. Instead, do the conversions once (at startup) and store the useful data types on the hashtable.

----------------------------------------------------------------------

Comment By: Chris Buccella (buccella)
Date: 2013-02-07 12:50

Message:
committed to HEAD

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603719 ] Conform to updated toolchain

[SourceForge.net <no...@so...>]

Bugs item #3603719, was opened at 2013-02-07 07:40
Message generated for change (Tracker Item Submitted) made by mchasal
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603719&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Metric Gatherer
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Michael Chase-Salerno (mchasal)
Assigned to: Tyrel Datwyler (tyreld)
Summary: Conform to updated toolchain

Initial Comment:
On the latest Ubuntu 12.10, I get the following errors during build:

libtool: link: gcc -Wall -g -O2 -o .libs/gatherctl gatherctl.o  ./.libs/librgather.so -lvirt -lsysfs -ldl -lpthread
/usr/bin/ld: gatherctl.o: undefined reference to symbol 'ch_release'
/usr/bin/ld: note: 'ch_release' is defined in DSO /home/msalerno/CODE/gather/.libs/libgatherutil.so.0 so try adding it to the linker command line
/home/msalerno/CODE/gather/.libs/libgatherutil.so.0: could not read symbols: Invalid operation
collect2: error: ld returned 1 exit status

I've run into this with other sblim projects as well and added some addition dependencies to address it, the attached patch cleans them up, but I understand that there may be a different approach.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603719&group_id=128809

[Sblim-issues] [ sblim-Bugs-3600125 ] Share client tracing shm segment

[SourceForge.net <no...@so...>]

Bugs item #3600125, was opened at 2013-01-09 09:59
Message generated for change (Settings changed) made by mchasal
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3600125&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcb
Group: Memory Leak
>Status: Pending
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Michael Chase-Salerno (mchasal)
Assigned to: Michael Chase-Salerno (mchasal)
Summary: Share client tracing shm segment

Initial Comment:
The fix to handle the shm leak under this bug: https://sourceforge.net/tracker/?func=detail&aid=3593006&group_id=128809&atid=712784 Doesn\'t address clients who create multiple cimc environments in a single process. This should be able to be handled by only creating a shm segment the first time tracing is initialized in a process and further calls would just reuse the same one. 

----------------------------------------------------------------------

Comment By: Michael Chase-Salerno (mchasal)
Date: 2013-01-09 10:20

Message:
pushed to git master

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3600125&group_id=128809

[Sblim-issues] [ sblim-Bugs-2284848 ] CPU utilization should account for steal-time

[SourceForge.net <no...@so...>]

Bugs item #2284848, was opened at 2008-11-14 08:22
Message generated for change (Comment added) made by mchasal
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=2284848&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Metric Gatherer
Group: Function
>Status: Pending
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: Narasimha Sharoff (nsharoff)
Assigned to: Michael Chase-Salerno (mchasal)
Summary: CPU utilization should account for steal-time

Initial Comment:
The Linux_ProcessorMetricValue reports this metrics.

The processor metrics is meant to have an OS view. Recent changes in Linux have introduced steal-time to account for the time where the LPAR took away the CPU. 

The SBLIM data gatherer code should be updated to
take account for this additional value.


----------------------------------------------------------------------

Comment By: Michael Chase-Salerno (mchasal)
Date: 2013-02-06 13:13

Message:
pushed to git master

----------------------------------------------------------------------

Comment By: Tyrel Datwyler (tyreld)
Date: 2008-12-02 12:52

Message:
The "ExternalView..." OperatingSystem metrics are the relevant metrics that
are effected by steal time.

----------------------------------------------------------------------

Comment By: Tyrel Datwyler (tyreld)
Date: 2008-12-02 12:44

Message:
This needs to be accounted for by both the Processor and OperatingSystem
metrics.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=2284848&group_id=128809

[Sblim-issues] [ sblim-Bugs-3601894 ] Enhance HTTP and CIM-XML tracing

[SourceForge.net <no...@so...>]

Bugs item #3601894, was opened at 2013-01-23 09:52
Message generated for change (Comment added) made by blaschke-oss
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3601894&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Java Client (JSR48)
Group: Usability
>Status: Pending
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Dave Blaschke (blaschke-oss)
Assigned to: Dave Blaschke (blaschke-oss)
Summary: Enhance HTTP and CIM-XML tracing

Initial Comment:
Currently, only HTTP headers from requests are being traced, so HTTP headers from responses, indications and trailers are not being traced.  Also, CIM-XML tracing does not distinguish between request and indication response (outgoing CIM-XML) as well as response and indication request (incoming CIM-XML).

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-02-06 06:58

Message:
The community review has completed and we received no substantial
criticism. Therefore the patch has been approved and merged into the "HEAD"
branch. The next release will pick it up.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-23 12:55

Message:
Patch sent for community review. During a 2 week period any
exploiter may comment on the patch, request changes or turn it
down completely (with good reason).  For the time being the patch is part
of the "Experimental" branch in CVS.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-23 10:14

Message:
With this patch, all HTTP headers can be found by grep'ing for "Headers="
with the following possible results:

Request HTTP Headers=
Response HTTP Headers=
Indication Request HTTP Headers=
Indication Response HTTP Headers=
Response HTTP Trailer Headers=
Indication Response HTTP Trailer Headers=
Indication Request HTTP Trailer Headers=

There is no need for "Request HTTP Trailer Headers" because the Java CIM
Client does not include trailers in its requests.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3601894&group_id=128809

[Sblim-issues] [ sblim-Bugs-3598613 ] different data type in cim instance and cim object path

[SourceForge.net <no...@so...>]

Bugs item #3598613, was opened at 2012-12-26 23:00
Message generated for change (Settings changed) made by blaschke-oss
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3598613&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Java Client (JSR48)
Group: Function
>Status: Pending
Resolution: Fixed
Priority: 5
Private: No
Submitted By: chen wang (wchencdl)
Assigned to: Dave Blaschke (blaschke-oss)
Summary: different data type in cim instance and cim object path

Initial Comment:
Steps to recreate
1. enumerate instances of PG_SSLCertificate in pg_internal
2. get the CloseableIterator instance
3. get the property of cim instance
4. get the property of the cim object path (it is returned by cimInstance.getObjectPath())

result:
data type in #3 is uint16, but data type in #4 is uint64

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-02-06 06:06

Message:
The community review has completed and we received no substantial
criticism. Therefore the patch has been approved and merged into the "HEAD"
branch. The next release will pick it up.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-17 06:41

Message:
Patch sent for community review. During a 2 week period any
exploiter may comment on the patch, request changes or turn it
down completely (with good reason).  For the time being the patch is part
of the "Experimental" branch in CVS.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-17 06:00

Message:
Option #2 was chosen, tested and shown to work in submitter's environment. 
The new property which enables/disables the new code is:

sblim.wbem.synchronizeNumericKeyDataTypes

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-14 05:36

Message:
Awaiting results of patch test by submitter

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-10 19:31

Message:
Forgot to mention that option #2 can also be controlled with property

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-10 13:20

Message:
The workaround for the CIM-XML spec deficiency is to go through each key in
the CIMObjectPath and, if it is a numeric data type, synchronize its data
type with the corresponding key in the CIMProperty[] (if it exists).  There
are two possibilities, based on where the key/property numeric data types
are synchronized:

1) CIMInstance

See attached file 3598613-patch-CIMInstance.txt.  This affects each and
every CIMInstance.  Main pro is every CIMInstance is covered, main con is
the performance impact.  Patch is simple (1 file).

2) CIM-XML Parsers

See attached file 3598613-patch-Parser.txt.  This affects only CIMInstances
in CIM-XML responses from CIMOM.  Main pro is better performance, main con
is only CIMInstances from CIMOMs are covered.  Patch is not as simple (5
files).

Please provide your feedback.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-09 07:04

Message:
What he is referring to by deficiency in the CIM-XML spec is that the TYPE
is required for a property:

<!ATTLIST PROPERTY %CIMName;
         %ClassOrigin;
         %Propagated;
         %EmbeddedObject;
         %CIMType;              #REQUIRED
         xml:lang   NMTOKEN     #IMPLIED
>

but not required for a key:

<!ATTLIST KEYVALUE VALUETYPE (string | boolean | numeric) "string"
        %CIMType;              #IMPLIED>
<!--

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-09 05:32

Message:
JSR48 chair response - 

"This is a deficiency in the CIM-XML specs - Other than Pegasus, SBLIM and
one proprietery implementation all other implementations (there are quite a
few, but I do not know how many support CIM-XML as many support only
WS-Man) support the optional type on keys that provides the real data type.
I have requested that SFCB support this on multiple occasions, but gave up
years ago when I received no response. We are deprecating the hint way, but
you will not be able to rely on it until 2.0. If we get an older server
that does not support the true type, we set it to CIMDataType.INVALID -
this is not an issue for when it gets to the provider since it knows the
type and the client can get the real type from the property as this only
effects the object path and not the properties. Clients should be treating
objecpaths as opaque.

"It would be nice if Pegasus and SFCB added support for the optional type.
I know that this is on the to-do list for Pegasus, but no clue when it will
be addressed."


----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-08 05:51

Message:
Note sent to JSR48 chair, awaiting response

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-07 14:38

Message:
That makes sense, as prior to 2.2.0 CIMInstance.getObjectPath() retrieved
an object path built by CIMInstance from the keys passed to its constructor
in both the CIMObjectPath and CIMProperty[] parameters, and the data type
from the key in CIMProperty[] overrides the data type from the key in
CIMObjectPath.

For example, consider the following code snippet:

CIMProperty<Short> p16 = new CIMProperty<Short>("AProp",
CIMDataType.SINT16_T, Short.valueOf("21"), true, false, null);
CIMProperty<Integer> p32 = new CIMProperty<Integer>("AProp",
CIMDataType.SINT32_T, Integer.valueOf("21"), true, false, null);
CIMProperty<?>[] a16 = new CIMProperty[] { p16 };
CIMProperty<?>[] a32 = new CIMProperty[] { p32 };
CIMObjectPath cop = new CIMObjectPath("http", "1.2.3.4", "5678", "doody",
"root/interop", a16);
CIMInstance i = new CIMInstance(cop, a32);
System.out.println("IT IS " + i.getProperty("AProp").getDataType() + " VS "
+ i.getObjectPath().getKey("AProp").getDataType(), false);

Output on 2.1.12: "IT IS sint32 VS sint32"
Output on 2.2.0 : "IT IS sint32 VS sint16"

----------------------------------------------------------------------

Comment By: Samuel (dingxg)
Date: 2013-01-05 18:42

Message:
yes, the code snippet we provided produced "unit16" in 2.1.x. 

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-02 10:54

Message:
This is more than likely a side effect of the 2.2.0 change which quit
merging the keys from the object path with the properties from the
instance.  Looking at the XML, you can see the key in the object path:

<KEYBINDING NAME="TruststoreType">
<KEYVALUE VALUETYPE="numeric">2</KEYVALUE>
</KEYBINDING>

is defined differently than the property in the instance:

<PROPERTY NAME="TruststoreType"  TYPE="uint16">
<VALUE>2</VALUE>
</PROPERTY>

"numeric" is generic and can mean uintXX or sintXX.  Further investigation
is required.

Did the code snipet you provided below produce "uint16" in 2.1.x?


            CIMInstance certificateInst = (CIMInstance)iter.next();
            System.out.println("Property type from CIM instance is " +
certificateInst.getProperty("truststoreType").getDataType());
            
            CIMObjectPath cimobjP = certificateInst.getObjectPath();
            System.out.println("Property type from CIM object path is " +
cimobjP.getKey("truststoreType").getDataType()

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2012-12-31 06:48

Message:
Apologies for the delay, the Java CIM Client development team is on
vacation and will return to work on January 2.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3598613&group_id=128809

[Sblim-issues] [ sblim-Bugs-3596303 ] windows http response WWW-Authenticate: Negotiate fails

[SourceForge.net <no...@so...>]

Bugs item #3596303, was opened at 2012-12-15 10:19
Message generated for change (Comment added) made by blaschke-oss
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3596303&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Java Client (JSR48)
Group: Security
>Status: Pending
Resolution: Fixed
Priority: 5
Private: No
Submitted By: nb950 (nb950)
Assigned to: Dave Blaschke (blaschke-oss)
Summary: windows  http response WWW-Authenticate: Negotiate fails

Initial Comment:
when server returns WWW-Authenticate: Negotiate  the sblim-cim-client2-2.1.7.jar does not responds with empty Authorization http header resulting in 401 
however if this Autheticate Negotiate is not presnt then the client sends the Authorization correct and it works

bad exchange :

WWW-Authenticate: Basic realm="CIM...@fo..."
WWW-Authenticate: Negotiate 
Content-Length: 0

RESPONSE :

Content-type: application/xml; charset="utf-8"
Content-Language: en-US
11-CIMMethod: EnumerateInstanceNames
Authorization: 
Content-length: 387

good exchange:

HTTP/1.1 401 Authorization Required
WWW-Authenticate: Basic realm="CIM...@fo..."
Content-Length: 0

M-POST /cimom HTTP/1.1
Connection: Keep-alive
11-CIMOperation: MethodCall
Content-type: application/xml; charset="utf-8"
Content-Language: en-US
11-CIMMethod: CreateInstance
Authorization: Basic xxxxxxxxxxxxxx==

notice the empry Authorization in the above bad example and  expected value in the good exchange  example above

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-02-06 05:04

Message:
The community review has completed and we received no substantial
criticism. Therefore the patch has been approved and merged into the "HEAD"
branch. The next release will pick it up.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-16 18:45

Message:
Patch sent for community review. During a 2 week period any
exploiter may comment on the patch, request changes or turn it
down completely (with good reason).  For the time being the patch is part
of the "Experimental" branch in CVS.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-16 15:42

Message:
Copyright dates will be updated by this bug

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-16 15:41

Message:
From submitter:

"I tested both patch's and it worked ok , please consider this issue fixed
thanks a lot , appreciate it"

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-14 05:35

Message:
Awaiting results of patch test by submitter

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-10 15:04

Message:
So if you are going to use this (or possibly use this) in a production
environment, then I'd prefer to do an official maintenance release.

The "current version" for the 2.1 code stream is 2.1.12.1, are you okay if
I apply the patches to that (2.1.12.2) and you upgrade to that, or are you
thinking 2.1.7?

I would like to test the new jar in your environment to see if the issue is
indeed fixed, can you help with that?  If so, email me (blaschke at
us.ibm.com) and I will send you a patched jar.  

----------------------------------------------------------------------

Comment By: nb950 (nb950)
Date: 2013-01-10 14:48

Message:
sorry for the delay , terrific turn around  thanks  i can use patch on the
current version , btw i am working at a startup

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-09 10:55

Message:
The fix for 3553858 is not a complete solution, as the Java CIM Client
still only parses the first challenge.  Another fix is needed (which I will
commit with this bug) that will enable the Client to parse all challenges.

The fix for 3553858 would actually work for you at the moment, as the first
challenge (Basic) is the one you need, but if the CIMOM decides to change
the ordering, i.e.:

WWW-Authenticate: Negotiate 
WWW-Authenticate: Basic realm="CIM...@fo..."

then you will be broken again.

I can provide a jar to test both patches, but again I need to know the
answers to the questions from my previous comment so I know where to
deliver it, if a maintenance release is required, etc.

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-04 08:25

Message:
Do you work for IBM?  Is this patch for personal (single use,
non-production environment, etc) or business (multiple use, production
environment, etc) use?  Need to know this info to determine delivery
options, thanks.

----------------------------------------------------------------------

Comment By: nb950 (nb950)
Date: 2013-01-03 17:00

Message:
>This issue is addressed in 2.2.0 by bug 3553858.
can you provide a fix in 2.1 branch ? appreciate it

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-03 03:21

Message:
The following is an example of a valid HTTP header field:

WWW-Authenticate: Basic realm="X1", Digest realm="X1",
domain="http://example.com", Basic realm="X2"

As mentioned in the previous comment, this makes for some interesting
parsing since there's three challenges and not four.  Now Windows comes
along and makes things even more interesting by adding something that seems
to go against the "challenge = auth-scheme 1*SP 1#auth-param" format since
Negotiate has a null auth-param, which makes the following possible:

WWW-Authenticate: Basic realm="X1", Digest realm="X1",
domain="http://example.com", Negotiate, Basic realm="X2"

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-02 06:55

Message:
It does not appear as though
org.sblim.cimclient.internal.http.Challenge.parseChallenge() handles
multiple challenges as defined by RFC2617:

      challenge   = auth-scheme 1*SP 1#auth-param
      auth-scheme    = token
      auth-param     = token "=" ( token | quoted-string )

   Note: User agents will need to take special care in parsing the WWW-
   Authenticate or Proxy-Authenticate header field value if it contains
   more than one challenge, or if more than one WWW-Authenticate header
   field is provided, since the contents of a challenge may itself
   contain a comma-separated list of authentication parameters.



----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2013-01-02 06:36

Message:
In 2.1.7, duplicate HTTP header fields overwrite the previous value.  In
your bad exchange example, the Java CIM Client ends up with:

WWW-Authenticate: Negotiate

and so basic authentication does not take place.  This issue is addressed
in 2.2.0 by bug 3553858.  In your bad exchange example, the Client now ends
up with:

WWW-Authenticate: Basic realm="CIM...@fo...", Negotiate

----------------------------------------------------------------------

Comment By: Dave Blaschke (blaschke-oss)
Date: 2012-12-31 06:47

Message:
Apologies for the delay, the Java CIM Client development team is on
vacation and will return to work on January 2.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3596303&group_id=128809

[Sblim-issues] [ sblim-Bugs-3529678 ] sfcc getClass does not expose method information

[SourceForge.net <no...@so...>]

Bugs item #3529678, was opened at 2012-05-25 00:39
Message generated for change (Comment added) made by hellerda
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcc
Group: New Feature
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Klaus Kämpf (kkaempf)
Assigned to: Dave Heller (hellerda)
Summary: sfcc getClass does not expose method information

Initial Comment:
Retrieving class information with getClass only exposes property and qualifier information but not method information.


----------------------------------------------------------------------

>Comment By: Dave Heller (hellerda)
Date: 2013-02-05 16:24

Message:
Patch for final review.  Adds support for class methods, including method
parameters and qualifiers.  Support for parameter references, qualifier
value arrays, deferred as future TODO.

----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-04 09:04

Message:
Preliminary patch attached.  This adds support for class methods, including
method parameters and qualifiers.  The program "TEST/test_gc" is expanded
to show the results.  TODO:
- Handle parameter references (currently they are parsed but not handled in
the parameter modeling)
- Handle qualifier value arrays (currently handles only values, not value
arrays)
- Add new error condition: CMPI_RC_ERR_NO_SUCH_METHOD
- Expand (and relocate?) utility function type2Chars()


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

[Sblim-issues] [ sblim-Bugs-3603454 ] control should do type conversions on startup

[SourceForge.net <no...@so...>]

Bugs item #3603454, was opened at 2013-02-05 12:31
Message generated for change (Tracker Item Submitted) made by buccella
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcb
Group: Performance
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Chris Buccella (buccella)
Assigned to: Chris Buccella (buccella)
Summary: control should do type conversions on startup

Initial Comment:
The control values (from sfcb.cfg) are stored in a hashtable, with the values being strings. Calls to getControlNum() will perform a strtol() call on the string value every time it is called. Similarly, getControlBool() will perform a strcmp() every time it is called. This is inefficient, since we never need to use those values as strings. Instead, do the conversions once (at startup) and store the useful data types on the hashtable.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3603454&group_id=128809

[Sblim-issues] [ sblim-Bugs-3529678 ] sfcc getClass does not expose method information

[SourceForge.net <no...@so...>]

Bugs item #3529678, was opened at 2012-05-25 00:39
Message generated for change (Comment added) made by hellerda
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcc
Group: New Feature
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Klaus Kämpf (kkaempf)
Assigned to: Dave Heller (hellerda)
Summary: sfcc getClass does not expose method information

Initial Comment:
Retrieving class information with getClass only exposes property and qualifier information but not method information.


----------------------------------------------------------------------

Comment By: Dave Heller (hellerda)
Date: 2013-02-04 09:04

Message:
Preliminary patch attached.  This adds support for class methods, including
method parameters and qualifiers.  The program "TEST/test_gc" is expanded
to show the results.  TODO:
- Handle parameter references (currently they are parsed but not handled in
the parameter modeling)
- Handle qualifier value arrays (currently handles only values, not value
arrays)
- Add new error condition: CMPI_RC_ERR_NO_SUCH_METHOD
- Expand (and relocate?) utility function type2Chars()


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

[Sblim-issues] [ sblim-Bugs-3529678 ] sfcc getClass does not expose method information

[SourceForge.net <no...@so...>]

Bugs item #3529678, was opened at 2012-05-25 00:39
Message generated for change (Settings changed) made by buccella
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: sfcc
Group: New Feature
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Klaus Kämpf (kkaempf)
>Assigned to: Dave Heller (hellerda)
Summary: sfcc getClass does not expose method information

Initial Comment:
Retrieving class information with getClass only exposes property and qualifier information but not method information.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=712784&aid=3529678&group_id=128809