[SOGo] Samba crashing

2014-06-19 Thread Rex Johnston
I'll just cut straight to the end here.

Running samba in the debugger...

Program received signal SIGSEGV, Segmentation fault.
ndr_push_mapi_response (ndr=0xd285b0, ndr_flags=optimized out, r=0x0)
at ndr_mapi.c:561
561 NDR_CHECK(ndr_push_uint16(ndr, NDR_SCALARS, r-length));
(gdb) bt
#0  ndr_push_mapi_response (ndr=0xd285b0, ndr_flags=optimized out,
r=0x0) at ndr_mapi.c:561
#1  0x7fffd8ac14f8 in dcesrv_EcDoRpcExt2 (r=0x13da110,
mem_ctx=0x307a9c0, dce_call=optimized out) at
mapiproxy/servers/default/emsmdb/dcesrv_exchange_emsmdb.c:1792
#2  dcesrv_exchange_emsmdb_dispatch (dce_call=optimized out,
mem_ctx=0x307a9c0, r=0x13da110, mapiproxy=optimized out) at
mapiproxy/servers/default/emsmdb/dcesrv_exchange_emsmdb.c:1943
#3  0x7fffd9c198fc in mapiproxy_server_dispatch (dce_call=0x307a9c0,
mem_ctx=0x307a9c0, r=0x13da110, mapiproxy=0x7fffdf10) at
mapiproxy/libmapiproxy/dcesrv_mapiproxy_server.c:57
#4  0x7fffd9e57b40 in mapiproxy_op_dispatch (dce_call=0x307a9c0,
mem_ctx=0x307a9c0, r=0x13da110) at mapiproxy/dcesrv_mapiproxy.c:504
#5  0x7fffe5cb9fd2 in dcesrv_request (call=0x307a9c0) at
../source4/rpc_server/dcerpc_server.c:965
#6  dcesrv_process_ncacn_packet (dce_conn=optimized out,
pkt=optimized out, blob=...) at ../source4/rpc_server/dcerpc_server.c:1110
#7  0x7fffe5cbab59 in dcesrv_read_fragment_done (subreq=0x0) at
../source4/rpc_server/dcerpc_server.c:1488
#8  0x74987698 in dcerpc_read_ncacn_packet_done (subreq=0x0) at
../librpc/rpc/dcerpc_util.c:295
#9  0x7151ac64 in tstream_readv_pdu_readv_done
(subreq=0x2bc4e40) at ../lib/tsocket/tsocket_helpers.c:319
#10 0x71519b64 in tstream_readv_done (subreq=0x0) at
../lib/tsocket/tsocket.c:604
#11 0x73f170e2 in tevent_common_loop_immediate () from
/usr/lib/x86_64-linux-gnu/libtevent.so.0
#12 0x73f19b26 in ?? () from
/usr/lib/x86_64-linux-gnu/libtevent.so.0
#13 0x73f16860 in _tevent_loop_once () from
/usr/lib/x86_64-linux-gnu/libtevent.so.0
#14 0x73f169eb in tevent_common_loop_wait () from
/usr/lib/x86_64-linux-gnu/libtevent.so.0
#15 0x00409459 in binary_smbd_main (argc=optimized out,
argv=optimized out, binary_name=0x40bf95 samba) at
../source4/smbd/server.c:486
#16 0x73b7476d in __libc_start_main () from
/lib/x86_64-linux-gnu/libc.so.6
#17 0x00405631 in _start ()
(gdb) print r
$1 = (const struct mapi_response *) 0x0


I've altered mapiproxy/servers/default/emsmdb/dcesrv_exchange_emsmdb.c:1782

if( !( mapi_response = EcDoRpc_process_transaction(mem_ctx,
emsmdbp_ctx, mapi2k7_request.mapi_request) ) )  // hmmm, Rex
return MAPI_E_INVALID_OBJECT;

It seems to have stopped the null pointer problem, but i have absolutely
no idea what it should do when EcDoRpc_process_transaction() returns NULL,
or even what MAPI_E_INVALID_OBJECT means.

Cheers, Rex

-- 
users@sogo.nu
https://inverse.ca/sogo/lists


Re: [SOGo] Samba crashing

2014-06-19 Thread Rex Johnston
Sorry, should have included that information.

Pulled from git on the 16th with git clone
https://github.com/openchange/openchange.git;

root@sogo:/usr/src/openchange_git/openchange# fgrep RELEASE VERSION
# e.g. OPENCHANGE_VERSION_RELEASE_NICKNAME=Nicky Nickname   #
OPENCHANGE_VERSION_RELEASE_NICKNAME=Nanoprobe
# e.g. OPENCHANGE_VERSION_RELEASE_VERSION=0.8.2 #
OPENCHANGE_VERSION_RELEASE_NUMBER=2.2


Cheers, Rex


Julien Kerihuel wrote:
 Hi Rex,

 What is the version of openchange you are experiencing this crash with?
 (git master, 3rd party repo?)

 Cheers,
 Julien.

 On 18/06/14 16:53, Rex Johnston wrote:
 I'll just cut straight to the end here.

 Running samba in the debugger...

 Program received signal SIGSEGV, Segmentation fault.
 ndr_push_mapi_response (ndr=0xd285b0, ndr_flags=optimized out, r=0x0)
 at ndr_mapi.c:561
 561 NDR_CHECK(ndr_push_uint16(ndr, NDR_SCALARS, r-length));
 (gdb) bt
 #0  ndr_push_mapi_response (ndr=0xd285b0, ndr_flags=optimized out,
 r=0x0) at ndr_mapi.c:561
 #1  0x7fffd8ac14f8 in dcesrv_EcDoRpcExt2 (r=0x13da110,
 mem_ctx=0x307a9c0, dce_call=optimized out) at
 mapiproxy/servers/default/emsmdb/dcesrv_exchange_emsmdb.c:1792
 #2  dcesrv_exchange_emsmdb_dispatch (dce_call=optimized out,
 mem_ctx=0x307a9c0, r=0x13da110, mapiproxy=optimized out) at
 mapiproxy/servers/default/emsmdb/dcesrv_exchange_emsmdb.c:1943
 #3  0x7fffd9c198fc in mapiproxy_server_dispatch (dce_call=0x307a9c0,
 mem_ctx=0x307a9c0, r=0x13da110, mapiproxy=0x7fffdf10) at
 mapiproxy/libmapiproxy/dcesrv_mapiproxy_server.c:57
 #4  0x7fffd9e57b40 in mapiproxy_op_dispatch (dce_call=0x307a9c0,
 mem_ctx=0x307a9c0, r=0x13da110) at mapiproxy/dcesrv_mapiproxy.c:504
 #5  0x7fffe5cb9fd2 in dcesrv_request (call=0x307a9c0) at
 ../source4/rpc_server/dcerpc_server.c:965
 #6  dcesrv_process_ncacn_packet (dce_conn=optimized out,
 pkt=optimized out, blob=...) at
../source4/rpc_server/dcerpc_server.c:1110
 #7  0x7fffe5cbab59 in dcesrv_read_fragment_done (subreq=0x0) at
 ../source4/rpc_server/dcerpc_server.c:1488
 #8  0x74987698 in dcerpc_read_ncacn_packet_done (subreq=0x0) at
 ../librpc/rpc/dcerpc_util.c:295
 #9  0x7151ac64 in tstream_readv_pdu_readv_done
 (subreq=0x2bc4e40) at ../lib/tsocket/tsocket_helpers.c:319
 #10 0x71519b64 in tstream_readv_done (subreq=0x0) at
 ../lib/tsocket/tsocket.c:604
 #11 0x73f170e2 in tevent_common_loop_immediate () from
 /usr/lib/x86_64-linux-gnu/libtevent.so.0
 #12 0x73f19b26 in ?? () from
 /usr/lib/x86_64-linux-gnu/libtevent.so.0
 #13 0x73f16860 in _tevent_loop_once () from
 /usr/lib/x86_64-linux-gnu/libtevent.so.0
 #14 0x73f169eb in tevent_common_loop_wait () from
 /usr/lib/x86_64-linux-gnu/libtevent.so.0
 #15 0x00409459 in binary_smbd_main (argc=optimized out,
 argv=optimized out, binary_name=0x40bf95 samba) at
 ../source4/smbd/server.c:486
 #16 0x73b7476d in __libc_start_main () from
 /lib/x86_64-linux-gnu/libc.so.6
 #17 0x00405631 in _start ()
 (gdb) print r
 $1 = (const struct mapi_response *) 0x0


 I've altered
mapiproxy/servers/default/emsmdb/dcesrv_exchange_emsmdb.c:1782

 if( !( mapi_response = EcDoRpc_process_transaction(mem_ctx,
 emsmdbp_ctx, mapi2k7_request.mapi_request) ) )  // hmmm, Rex
 return MAPI_E_INVALID_OBJECT;

 It seems to have stopped the null pointer problem, but i have absolutely
 no idea what it should do when EcDoRpc_process_transaction() returns
NULL,
 or even what MAPI_E_INVALID_OBJECT means.

 Cheers, Rex




-- 
users@sogo.nu
https://inverse.ca/sogo/lists


[SOGo] Scalix to SOGo migration...

2014-05-14 Thread Rex Johnston
Hi Guys,

I have 2 issues left here, in this migration.

1) Outlook won't automatically display new messages.  I see someone else
had this issue over a year ago.  I find it hard to believe such an issue
still exists...

2) Error handling in Openchange/SOGo.  The data imported from Scalix is,
to be frank, a bit of a mess.  Non standard extensions, corruption etc. 
Hey, it's one of the
reasons it's being replaced.  Anyway.  I'm trying to fix up the
openchange and sogo code such that, when an error occurs, the whole
openchange plugin doesn't exit
for the whole office.  This isn't the way to handle poor quality data
fellas.  Objective C (SOGo) isn't something i've used before, but i'm
happy hacking away on the
openchange end.  Anyway, i've made a few changes, to stop the server
abort()'ing. How does one contribute?

Cheers, Rex
-- 
users@sogo.nu
https://inverse.ca/sogo/lists


[SOGo] Outlook Sync issues.

2014-04-10 Thread Rex Johnston
Hi,

I'm (hopefully) in the final stages of deploying SOGo/Openchange and i'm
running in to a few Outlook issues.
After upgrading to SOGo 2.2.3-1 i've fixed most problems, but now get an
Outlook sync issue.

Hitting send receive, i get 0X8004010F object not found blah blah. 
Looking at the sync log, it tells me

Microsoft Exchange offline address book
Not downloading Offline address book files. A server (URL) could not be
located.
0X8004010F

Random sampling of log.samba doesn't tell me anything, no obvious
references to useful errors.

Can anyone point me in the right direction?

Thanks, Rex
-- 
users@sogo.nu
https://inverse.ca/sogo/lists


Re: [SOGo] Unable to compile current git archive

2013-07-23 Thread Rex Johnston

Ludovic Marcotte wrote:

On 2013-07-22 7:02 PM, r...@sclnz.com wrote:

Can anyone help?

Try this patch:

index 6ad78ee..ced651d 100644
--- a/UI/GNUmakefile
+++ b/UI/GNUmakefile
@@ -8,8 +8,8 @@ SUBPROJECTS += \
SOGoElements\
Common  \
Contacts\
+   MailPartViewers \
MailerUI\
-   MailPartViewers \
MainUI  \
PreferencesUI   \
Scheduler   \



Ah i see.  We might have a bit of a dependency loop?

Making all in Contacts...
Making all for bundle ContactsUI...
make[4]: Nothing to be done for `internal-bundle-compile'.
 Copying resources into the bundle wrapper...
 Copying localized resources into the bundle wrapper...
Making all in MailPartViewers...
Making all for bundle MailPartViewers...
 Compiling file ../../SoObjects/Mailer/SOGoMailBodyPart.m ...
../../SoObjects/Mailer/SOGoMailBodyPart.m:118:2: warning: #warning 
partToPart should be populated directly
../../SoObjects/Mailer/SOGoMailBodyPart.m:507: fatal error: opening 
dependency file obj/../../SoObjects/Mailer/SOGoMailBodyPart.m.d: No such 
file or directory

compilation terminated.
make[4]: *** [obj/../../SoObjects/Mailer/SOGoMailBodyPart.m.o] Error 1
make[3]: *** [internal-bundle-run-compile-submake] Error 2
make[2]: *** [MailPartViewers.all.bundle.variables] Error 2
make[1]: *** [internal-all] Error 2
make: *** [internal-all] Error 2

Cheers, Rex

--
users@sogo.nu
https://inverse.ca/sogo/lists


Re: [SOGo] Unable to compile current git archive

2013-07-23 Thread Rex Johnston

Ludovic Marcotte wrote:

The real fix was committed a few hours ago.


Thanks, SOGo compiled OK.

Onto OpenChange, i get this...

root@rimu:~/src/openchange/sogo# cd OpenChange/
root@rimu:~/src/openchange/sogo/OpenChange# make
This is gnustep-make 2.2.0. Type 'make print-gnustep-make-help' for help.
 Python executable: /usr/bin/python
 Auto-generating MAPIStorePropertySelectors.[hm]...
Making all for bundle SOGoBackend...
 Creating SOGoBackend.MAPIStore/
 Compiling file MAPIApplication.m ...
 Compiling file MAPIStoreActiveTables.m ...
 Compiling file MAPIStoreAuthenticator.m ...
 Compiling file MAPIStoreMapping.m ...
 Compiling file MAPIStoreMIME.m ...
snip
 Compiling file iCalTimeZone+MAPIStore.m ...
 Compiling file GCSSpecialQueries+OpenChange.m ...
 Compiling file EOQualifier+MAPI.m ...
 Compiling file EOBitmaskQualifier.m ...
 Compiling file BSONCodec.m ...
cc1obj: warnings being treated as errors
BSONCodec.m: In function ‘-[NSObject(BSONObjectCoding) BSONEncode]’:
BSONCodec.m:88: error: implicit declaration of function 
‘class_conformsToProtocol’

BSONCodec.m:94: error: implicit declaration of function ‘class_getName’
BSONCodec.m:94: error: initialization makes pointer from integer without 
a cast

make[2]: *** [obj/BSONCodec.m.o] Error 1
make[1]: *** [internal-bundle-run-compile-submake] Error 2
make: *** [SOGoBackend.all.bundle.variables] Error 2
root@rimu:~/src/openchange/sogo/OpenChange#

Is it safe to get the compiler to ignore warnings?

Many thanks.

Rex
--
users@sogo.nu
https://inverse.ca/sogo/lists