brutus server and evoution connector

[Thomas Thompson]

Hello all,

First off, many thanks to Jules for creating and maintaining a MAPI solution.

I've been trying to get the brutus server and evolution connector to work properly, and I'm running into a bit of a snag. I obtained the latest brutus server from the download link provided to the list (2.0.6 according to the log), and am using the connector version 1.2.35-2 from the Fedora 11 repository. I'm not sure if the connector in the fedora repositories is built with debugging option enabled (I added a .brutus-do-debug file in my home directory and did not receive debug outputs when running evolution and setting up the brutus account). In any case, the connector asks for my pasword when first enabling the account. From that point on, if I try to do anything with the account in evolution (like expand the folder tree for example), it prompts me for a password again without takin
g the action. No errors are reported through evolution.

On the server side, the log file seems to indicate everything started correctly on the server, and I can see logs were my client attempts to connect:

< 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:708 - Entering BRUTUS_BrutusLogOn_i::GetVersion()
< 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:726 - Leaving BRUTUS_BrutusLogOn_i::GetVersion()
< 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:708 - Entering BRUTUS_BrutusLogOn_i::GetVersion()
< 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:726 - Leaving BRUTUS_BrutusLogOn_i::GetVersion()

Is there a version mismatch between the server and the client (should I be using an older brutus server build or upgrade the connector)?

----------------------
Thomas Thompson
Senior Software Engineer
Mechdyne Corporation
757-456-5500 ext 6110

[Jules Colding]

On 16/06/2009, at 20.13, Thomas Thompson wrote:

> Hello all,
>
> First off, many thanks to Jules for creating and maintaining a MAPI
> solution.

Thanks :-)


> I've been trying to get the brutus server and evolution connector to
> work properly, and I'm running into a bit of a snag. I obtained
> the latest brutus server from the download link provided to the list
> (2.0.6 according to the log), and am using the connector version
> 1.2.35-2 from the Fedora 11 repository. I'm not sure if the
> connector in the fedora repositories is built with debugging option
> enabled (I added a .brutus-do-debug file in my home directory and
> did not receive debug outputs when running evolution and setting up
> the brutus account). In any case, the connector asks for my pasword
> when first enabling the account. From that point on, if I try to do
> anything with the account in evolution (like expand the folder tree
> for example), it prompts me for a password again without taking the
> action. No errors are reported through evolution.
>
> On the server side, the log file seems to indicate everything
> started correctly on the server, and I can see logs were my client
> attempts to connect:
>
> < 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:708 - Entering
> BRUTUS_BrutusLogOn_i::GetVersion()
> < 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:726 - Leaving
> BRUTUS_BrutusLogOn_i::GetVersion()
> < 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:708 - Entering
> BRUTUS_BrutusLogOn_i::GetVersion()
> < 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:726 - Leaving
> BRUTUS_BrutusLogOn_i::GetVersion()
>
> Is there a version mismatch between the server and the client

I'm sure there are. The yum version is rather old compared to what is
in svn.


> (should I be using an older brutus server build or upgrade the
> connector)?

You should build from svn instead:

svn co http://svn.42tools.com/repos/evolution-brutus/trunk/ e-b

Best,
jules

[Thomas Thompson]

Thanks for the response Jules.

I think portions of the svn tree are out of date, here is the list of changes I had to make to get the IDL compile completed:

• idl/Makefile Change .net to .com for repository locaitons
• idl/Makefile added AsyncSession.idl (Required by BrutusLogin)
• idl_output/Makefile Add AsyncSessionCB-common.lo AsyncSessionCB-stubs.lo
• idl_outpit/Makefile Add AsyncSessionCB.idl to IDL file list.

After making these changes, the IDL compilation now works, however I'm getting the following compilation errors before the make fails:

brutus_util.c: In function ‘brutus_logon_loop’:
brutus_util.c:777: error: passing argument 3 of ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
../idl_output/BrutusLogOn.h:168: note: expected ‘BRUTUS_AsyncSessionCB’ but argument is of type ‘const char *’
brutus_util.c:777: error: passing argument 10 of ‘BRUTUS_BrutusLogOn_Logon’ makes pointer from integer without a cast
../idl_output/BrutusLogOn.h:168: note: expected ‘const CORBA_char *’ but argument is of type ‘unsigned int’
brutus_util.c:777: error: incompatible type for argument 13 of ‘BRUTUS_BrutusLogOn_Logon’
../idl_output/BrutusLogOn.h:168: note: expected ‘BRUTUS_SERVER_TYPE’ but argument is of type ‘CORBA_unsigned_long *’
brutus_util.c:777: error: passing argument 14 of ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
../idl_output/BrutusLogOn.h:168: note: expected ‘struct CORBA_Environment *’ but argument is of type ‘struct CORBA_Object_type **’
brutus_util.c:777: error: too many arguments to function ‘BRUTUS_BrutusLogOn_Logon’
brutus_util.c:866: error: passing argument 3 of ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
../idl_output/BrutusLogOn.h:168: note: expected ‘BRUTUS_AsyncSessionCB’ but argument is of type ‘const char *’
brutus_util.c:866: error: passing argument 10 of ‘BRUTUS_BrutusLogOn_Logon’ makes pointer from integer without a cast
../idl_output/BrutusLogOn.h:168: note: expected ‘const CORBA_char *’ but argument is of type ‘unsigned int’
brutus_util.c:866: error: incompatible type for argument 13 of ‘BRUTUS_BrutusLogOn_Logon’
../idl_output/BrutusLogOn.h:168: note: expected ‘BRUTUS_SERVER_TYPE’ but argument is of type ‘CORBA_unsigned_long *’
brutus_util.c:866: error: passing argument 14 of ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
../idl_output/BrutusLogOn.h:168: note: expected ‘struct CORBA_Environment *’ but argument is of type ‘struct CORBA_Object_type **’
brutus_util.c:866: error: too many arguments to function ‘BRUTUS_BrutusLogOn_Logon’
make[3]: *** [brutus_util.lo] Error 1

Thanks again for your help.

----------------------
Thomas Thompson
Senior Software Engineer
Mechdyne Corporation

________________________________________
From: brutus-bounces-***@public.gmane.org [brutus-bounces-***@public.gmane.org] On Behalf Of Jules Colding [colding-***@public.gmane.org]
Sent: Tuesday, June 16, 2009 3:45 PM
To: brutus-list
Subject: Re: brutus server and evoution connector

On 16/06/2009, at 20.13, Thomas Thompson wrote:

> Hello all,
>
> First off, many thanks to Jules for creating and maintaining a MAPI
> solution.

Thanks :-)


> I've been trying to get the brutus server and evolution connector to
> work properly, and I'm running into a bit of a snag. I obtained
> the latest brutus server from the download link provided to the list
> (2.0.6 according to the log), and am using the connector version
> 1.2.35-2 from the Fedora 11 repository. I'm not sure if the
> connector in the fedora repositories is built with debugging option
> enabled (I added a .brutus-do-debug file in my home directory and
> did not receive debug outputs when running evolution and setting up
> the brutus account). In any case, the connector asks for my pasword
> when first enabling the account. From that point on, if I try to do
> anything with the account in evolution (like expand the folder tree
> for example), it prompts me for a password again without taking the
> action. No errors are reported through evolution.
>
> On the server side, the log file seems to indicate everything
> started correctly on the server, and I can see logs were my client
> attempts to connect:
>
> < 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:708 - Entering
> BRUTUS_BrutusLogOn_i::GetVersion()
> < 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:726 - Leaving
> BRUTUS_BrutusLogOn_i::GetVersion()
> < 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:708 - Entering
> BRUTUS_BrutusLogOn_i::GetVersion()
> < 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:726 - Leaving
> BRUTUS_BrutusLogOn_i::GetVersion()
>
> Is there a version mismatch between the server and the client

I'm sure there are. The yum version is rather old compared to what is
in svn.


> (should I be using an older brutus server build or upgrade the
> connector)?

You should build from svn instead:

svn co http://svn.42tools.com/repos/evolution-brutus/trunk/ e-b

Best,
jules

[Jules Colding]

Hi Thomas,


On 18/06/2009, at 15.45, Thomas Thompson wrote:
>
> I think portions of the svn tree are out of date,

Yes, they are and what you've got won't work. It should be fixed now.
Please 'svn up' your local copy.

Thanks for letting me know.

Best,
jules




> here is the list of changes I had to make to get the IDL compile
> completed:
>
> • idl/Makefile Change .net to .com for repository locaitons
> • idl/Makefile added AsyncSession.idl (Required by BrutusLogin)
> • idl_output/Makefile Add AsyncSessionCB-common.lo AsyncSessionCB-
> stubs.lo
> • idl_outpit/Makefile Add AsyncSessionCB.idl to IDL file list.
>
> After making these changes, the IDL compilation now works, however
> I'm getting the following compilation errors before the make fails:
>
> brutus_util.c: In function ‘brutus_logon_loop’:
> brutus_util.c:777: error: passing argument 3 of
> ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
> ../idl_output/BrutusLogOn.h:168: note: expected
> ‘BRUTUS_AsyncSessionCB’ but argument is of type ‘const char *’
> brutus_util.c:777: error: passing argument 10 of
> ‘BRUTUS_BrutusLogOn_Logon’ makes pointer from integer without a cast
> ../idl_output/BrutusLogOn.h:168: note: expected ‘const CORBA_char *’
> but argument is of type ‘unsigned int’
> brutus_util.c:777: error: incompatible type for argument 13 of
> ‘BRUTUS_BrutusLogOn_Logon’
> ../idl_output/BrutusLogOn.h:168: note: expected ‘BRUTUS_SERVER_TYPE’
> but argument is of type ‘CORBA_unsigned_long *’
> brutus_util.c:777: error: passing argument 14 of
> ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
> ../idl_output/BrutusLogOn.h:168: note: expected ‘struct
> CORBA_Environment *’ but argument is of type ‘struct
> CORBA_Object_type **’
> brutus_util.c:777: error: too many arguments to function
> ‘BRUTUS_BrutusLogOn_Logon’
> brutus_util.c:866: error: passing argument 3 of
> ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
> ../idl_output/BrutusLogOn.h:168: note: expected
> ‘BRUTUS_AsyncSessionCB’ but argument is of type ‘const char *’
> brutus_util.c:866: error: passing argument 10 of
> ‘BRUTUS_BrutusLogOn_Logon’ makes pointer from integer without a cast
> ../idl_output/BrutusLogOn.h:168: note: expected ‘const CORBA_char *’
> but argument is of type ‘unsigned int’
> brutus_util.c:866: error: incompatible type for argument 13 of
> ‘BRUTUS_BrutusLogOn_Logon’
> ../idl_output/BrutusLogOn.h:168: note: expected ‘BRUTUS_SERVER_TYPE’
> but argument is of type ‘CORBA_unsigned_long *’
> brutus_util.c:866: error: passing argument 14 of
> ‘BRUTUS_BrutusLogOn_Logon’ from incompatible pointer type
> ../idl_output/BrutusLogOn.h:168: note: expected ‘struct
> CORBA_Environment *’ but argument is of type ‘struct
> CORBA_Object_type **’
> brutus_util.c:866: error: too many arguments to function
> ‘BRUTUS_BrutusLogOn_Logon’
> make[3]: *** [brutus_util.lo] Error 1
>
> Thanks again for your help.
>
> ----------------------
> Thomas Thompson
> Senior Software Engineer
> Mechdyne Corporation
>
> ________________________________________
> From: brutus-bounces-***@public.gmane.org [brutus-bounces-***@public.gmane.org] On
> Behalf Of Jules Colding [colding-***@public.gmane.org]
> Sent: Tuesday, June 16, 2009 3:45 PM
> To: brutus-list
> Subject: Re: brutus server and evoution connector
>
> On 16/06/2009, at 20.13, Thomas Thompson wrote:
>
>> Hello all,
>>
>> First off, many thanks to Jules for creating and maintaining a MAPI
>> solution.
>
> Thanks :-)
>
>
>> I've been trying to get the brutus server and evolution connector to
>> work properly, and I'm running into a bit of a snag. I obtained
>> the latest brutus server from the download link provided to the list
>> (2.0.6 according to the log), and am using the connector version
>> 1.2.35-2 from the Fedora 11 repository. I'm not sure if the
>> connector in the fedora repositories is built with debugging option
>> enabled (I added a .brutus-do-debug file in my home directory and
>> did not receive debug outputs when running evolution and setting up
>> the brutus account). In any case, the connector asks for my pasword
>> when first enabling the account. From that point on, if I try to do
>> anything with the account in evolution (like expand the folder tree
>> for example), it prompts me for a password again without taking the
>> action. No errors are reported through evolution.
>>
>> On the server side, the log file seems to indicate everything
>> started correctly on the server, and I can see logs were my client
>> attempts to connect:
>>
>> < 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:708 - Entering
>> BRUTUS_BrutusLogOn_i::GetVersion()
>> < 10:31:32.750000 > .\BrutusLogOnS_impl.cpp:726 - Leaving
>> BRUTUS_BrutusLogOn_i::GetVersion()
>> < 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:708 - Entering
>> BRUTUS_BrutusLogOn_i::GetVersion()
>> < 10:31:57.625000 > .\BrutusLogOnS_impl.cpp:726 - Leaving
>> BRUTUS_BrutusLogOn_i::GetVersion()
>>
>> Is there a version mismatch between the server and the client
>
> I'm sure there are. The yum version is rather old compared to what is
> in svn.
>
>
>> (should I be using an older brutus server build or upgrade the
>> connector)?
>
> You should build from svn instead:
>
> svn co http://svn.42tools.com/repos/evolution-brutus/trunk/ e-b
>
> Best,
> jules
>
>
> _______________________________________________
> brutus mailing list
> brutus-***@public.gmane.org
> http://www.42tools.com/mailman/listinfo/brutus
> _______________________________________________
> brutus mailing list
> brutus-***@public.gmane.org
> http://www.42tools.com/mailman/listinfo/brutus

[Jules Colding]

On 18/06/2009, at 15.56, Jules Colding wrote:

> Hi Thomas,
>
>
> On 18/06/2009, at 15.45, Thomas Thompson wrote:
>>
>> I think portions of the svn tree are out of date,
>
> Yes, they are and what you've got won't work. It should be fixed now.
> Please 'svn up' your local copy.

BTW, You may need to run "make distclean" to remove the non-working
idl files.

Thanks,
jules

[Thomas Thompson]

Jules,

Thank for the quick response. I restarted the autogen.sh / make on a clean svn checkout and everything built fine.

Now I'm back to my original issue where the server access fails after I provide my password. As an additional test, I tried to send an e-mail. Sending failed with evolution providing an "cannot get folder: invalid operation on this store" error.

Is there an option that should be specified to autogen to turn on debugging? The server log reamins much the same as the before (perhaps I should build the server too, the only problem I have is a lack of a dev environment on the windows side...)

For the e-b settings, I'm specifying port 4082 as the connection port, rather than the default 2003 in the client. The server log just shows the ge version enter / exit as mailed previously.

----------------------
Thomas Thompson
Senior Software Engineer
Mechdyne Corporation
757-456-5500 ext 6110
________________________________________
From: brutus-bounces-***@public.gmane.org [brutus-bounces-***@public.gmane.org] On Behalf Of Jules Colding [colding-***@public.gmane.org]
Sent: Thursday, June 18, 2009 10:01 AM
To: brutus-list
Subject: Re: brutus server and evoution connector

On 18/06/2009, at 15.56, Jules Colding wrote:

> Hi Thomas,
>
>
> On 18/06/2009, at 15.45, Thomas Thompson wrote:
>>
>> I think portions of the svn tree are out of date,
>
> Yes, they are and what you've got won't work. It should be fixed now.
> Please 'svn up' your local copy.

BTW, You may need to run "make distclean" to remove the non-working
idl files.

Thanks,
jules

[Jules Colding]

On 18/06/2009, at 19.11, Thomas Thompson wrote:

> Jules,
>
> Thank for the quick response. I restarted the autogen.sh / make on
> a clean svn checkout and everything built fine.
>
> Now I'm back to my original issue where the server access fails
> after I provide my password. As an additional test, I tried to send
> an e-mail. Sending failed with evolution providing an "cannot get
> folder: invalid operation on this store" error.
>
> Is there an option that should be specified to autogen to turn on
> debugging?

--enable-brutus-debug


> The server log reamins much the same as the before (perhaps I should
> build the server too, the only problem I have is a lack of a dev
> environment on the windows side...)

Please don't bother building the server unless you enjoy pain :-) I'm
sure that the e-b log will give you a clue as to what is wrong.


>
> For the e-b settings, I'm specifying port 4082 as the connection
> port, rather than the default 2003 in the client. The server log
> just shows the ge version enter / exit as mailed previously.

So the server works as it should. Do you use lorica on the client
side? As far as I remember that is more or less a requirement by now.
Please use this tag:

* http://svn.42tools.com/repos/lorica/tags/release-1.0.13.1


One more thing - I'm on my way to France now and I'll be gone for a
full 11 days, so please by don't expect an answer to any more mails
until I'm back again.

Best,
jules




>
> ----------------------
> Thomas Thompson
> Senior Software Engineer
> Mechdyne Corporation
> 757-456-5500 ext 6110
> ________________________________________
> From: brutus-bounces-***@public.gmane.org [brutus-bounces-***@public.gmane.org] On
> Behalf Of Jules Colding [colding-***@public.gmane.org]
> Sent: Thursday, June 18, 2009 10:01 AM
> To: brutus-list
> Subject: Re: brutus server and evoution connector
>
> On 18/06/2009, at 15.56, Jules Colding wrote:
>
>> Hi Thomas,
>>
>>
>> On 18/06/2009, at 15.45, Thomas Thompson wrote:
>>>
>>> I think portions of the svn tree are out of date,
>>
>> Yes, they are and what you've got won't work. It should be fixed now.
>> Please 'svn up' your local copy.
>
> BTW, You may need to run "make distclean" to remove the non-working
> idl files.
>
> Thanks,
> jules
>
>
>
>
>
> _______________________________________________
> brutus mailing list
> brutus-***@public.gmane.org
> http://www.42tools.com/mailman/listinfo/brutus
> _______________________________________________
> brutus mailing list
> brutus-***@public.gmane.org
> http://www.42tools.com/mailman/listinfo/brutus