Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-07 Thread Paul Smith
On Thu, 2009-12-03 at 22:34 +0100, Thomas Novin wrote:
 On Mon, 2009-11-30 at 14:42 -0500, Paul Smith wrote:
  Hi all.  I'm really confused by messages from people who say they're
  using Evolution with MAPI support and it's working just fine for them.
  I can't understand it: it's so far from working for me that there must
  be something I'm doing wrong or something about my environment which is
  very different from others.
 
 I tried evolution-mapi 2.28.1 in Ubuntu Karmic 32-bit (read bug
 https://bugs.launchpad.net/bugs/472552 about getting 2.28.1 in karmic).
 
 Except that all my calendar entries are off by one hour, they are one
 hour early I actually haven't found one bug. I have read lots of emails
 in different folder, looked at calendar entries back and forth.

I tried this and it's a disaster.  Every single attempt to connect to
the Exchange 2007 server causes Evolution to dump core.  I had to start
it with --offline to keep it up long enough that I could delete my
Exchange MAPI account.

I then tried to re-add my Exchange MAPI account and the instant I
clicked the Authenticate button in the add new account wizard,
Evolution dumped core again.

At least with the latest code on the gnome-2.28 branch (built from git)
Evo doesn't crash.  Of course I still have tons of bugs, but it stays
up! :-)

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-07 Thread Paul Smith
On Mon, 2009-12-07 at 14:25 -0500, Paul Smith wrote:
 I tried this and it's a disaster.  Every single attempt to connect to
 the Exchange 2007 server causes Evolution to dump core.  I had to start
 it with --offline to keep it up long enough that I could delete my
 Exchange MAPI account.

I vaguely remembered that you have to use the Exchange server IP
address, not hostname (lame!!) so I tried that and I did get it to
connect without crashing this time (uber-lame!!)

However, I still see some of the same problems as before: about half of
my inbox has no subject line listed in the summary window.  The calendar
does seem to work better but there are still a number of meetings
missing that should be there.  I can't query free/busy information on
other attendees when I create meetings (critically important!)  I
haven't tried things like sending meeting invites.

Any email I send to external addresses still has the TNEF attachment.

GAL seems to actually work with this, though, which is nice!

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-07 Thread Johnny Jacob
On Mon, 2009-12-07 at 14:39 -0500, Paul Smith wrote:
 On Mon, 2009-12-07 at 14:25 -0500, Paul Smith wrote:
  I tried this and it's a disaster.  Every single attempt to connect
 to
  the Exchange 2007 server causes Evolution to dump core.  I had to
 start
  it with --offline to keep it up long enough that I could delete my
  Exchange MAPI account.
 
 I vaguely remembered that you have to use the Exchange server IP
 address, not hostname (lame!!) so I tried that and I did get it to
 connect without crashing this time (uber-lame!!)

This crash seems to be a issue with the specific distro builds. In suse,
this was solved with compiler flags such as -Bsymbolic (nasty!)

Since we are using samba4 for libmapi and most of the distros ship
samba3, there were namespace collision caused by some krb5_* symbols.

 
 However, I still see some of the same problems as before: about half
 of
 my inbox has no subject line listed in the summary window.  The
 calendar
 does seem to work better but there are still a number of meetings
 missing that should be there.  I can't query free/busy information on
 other attendees when I create meetings (critically important!)  I
 haven't tried things like sending meeting invites.
 
 Any email I send to external addresses still has the TNEF attachment.
 
 GAL seems to actually work with this, though, which is nice!
 
 ___
 Evolution-hackers mailing list
 Evolution-hackers@gnome.org
 http://mail.gnome.org/mailman/listinfo/evolution-hackers


___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-07 Thread Paul Smith
On Tue, 2009-12-08 at 08:47 +0530, Johnny Jacob wrote:
  I vaguely remembered that you have to use the Exchange server IP
  address, not hostname (lame!!) so I tried that and I did get it to
  connect without crashing this time (uber-lame!!)
 
 This crash seems to be a issue with the specific distro builds. In
 suse, this was solved with compiler flags such as -Bsymbolic (nasty!)

Odd.  If I build Evo myself from source (latest gnome-2.28 git branch
and/or latest master git branch) I don't have this problem: using the
FQDN works just fine (that's why I'd forgotten about needing to do
this).

In addition to Evo and its parts, I'm also compiling gtkhtml, libsoup,
libxml2, and openchange from source.

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-04 Thread Milan Crha
On Thu, 2009-12-03 at 16:10 -0500, Reid Thompson wrote:
 hmm -- is calling evolution supposed to auto-start these two?

On Thu, 2009-12-03 at 16:12 -0500, Reid Thompson wrote:
 (evolution:24437): libebook-WARNING
 **: ../../../../evolution-data-server/addressbook/libebook/e-book.c:2194: 
 cannot activate book: The name org.gnome.evolution.dataserver.AddressBook was 
 not provided by any .service files 

Hi,
yes, those two processes are run as a DBus services. The relevant files
are in your $PREFIX/share/dbus-1/services/ and dbus is supposed to be
directed to that directory too, to know about them.

There are more options to do that, what I do is making a symlink to that
folder:
   $ rm /usr/local/share/dbus-1 2/dev/null
   $ ln -s $PREFIX/share/dbus-1 /usr/local/share/dbus-1
The only disadvantage of this solution is to have changed rights
on /usr/local/share/ to be able to write there even as a regular user.
Bye,
Milan

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-04 Thread Ross Burton
On Thu, 2009-12-03 at 16:12 -0500, Reid Thompson wrote:
 (evolution:24437): libebook-WARNING
 **: ../../../../evolution-data-server/addressbook/libebook/e-book.c:2194: 
 cannot activate book: The name org.gnome.evolution.dataserver.AddressBook was 
 not provided by any .service files 

You installed evolution-data-server into a prefix that DBus doesn't know
about, so it can't autostart the daemons.

I have this in my /etc/dbus-1/session.conf:

  servicedir/home/ross/BUILD/share/dbus-1/services//servicedir

Ross
-- 
Ross Burton mail: r...@burtonini.com
  jabber: r...@burtonini.com
   www: http://burtonini.com


signature.asc
Description: This is a digitally signed message part
___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-04 Thread Paul Smith
On Fri, 2009-12-04 at 09:30 +, Ross Burton wrote:
 You installed evolution-data-server into a prefix that DBus doesn't
 know about, so it can't autostart the daemons. 

Huh.  Well, that could definitely be a major part of my problem :-)

The question is, isn't there any way to provide a local configuration to
d-bus, similar to the BONOBO_ACTIVATION_PATH in bonobo?

I looked and it seems that there's a local user dbus-daemon that's
started, but it still reads the system session.conf file.  And I looked
in the session.conf file and it includes session-local.conf which is
supposed to be what you customize, if you need to customize dbus
locally... but that file appears to be defined to live in /etc/dbus-1
and so it's not REALLY a per-user customizable file.

Is there nothing in dbus that lets the user configure things, without
requiring root privileges?

Thanks!

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-04 Thread Reid Thompson
On Fri, 2009-12-04 at 09:30 +, Ross Burton wrote:
 On Thu, 2009-12-03 at 16:12 -0500, Reid Thompson wrote:
  (evolution:24437): libebook-WARNING
  **:
 ../../../../evolution-data-server/addressbook/libebook/e-book.c:2194:
 cannot activate book: The name
 org.gnome.evolution.dataserver.AddressBook was not provided by any
 .service files 
 
 You installed evolution-data-server into a prefix that DBus doesn't
 know
 about, so it can't autostart the daemons.
 
 I have this in my /etc/dbus-1/session.conf:
 
   servicedir/home/ross/BUILD/share/dbus-1/services//servicedir
 
 Ross

I still have something a bit off...

I added   servicedir/opt/evo/share/dbus-1/services//servicedir
as shown below.  The calendar factory and the addressbook factory now
start.

Any idea why  /usr/libexec/evolution-data-server-2.28 
--oaf-activate-iid=OAFIID:GNOME_Evolution_DataServer_BookFactory:1.2 
--oaf-ior-fd=29
would be getting started up?


calendar - populates using Exchange OWA
 - does not populate using MAPI

Addressbook - personal - works
- Exchange LDAP - works
- OWA Contacts - works
- OWA Contacts backup - works
- OWA GAL - Error loading address book.
This address book cannot be opened.
This either means that an incorrect URI
was entered, or the server is
unreachable.
   This was working in 2.28 ( perhaps I typoed something )

   - MAPI Contacts, Contacts backup, GAL - I get no response
of any type from any of the three.



$ cat /etc/dbus-1/session.conf 
!-- This configuration file controls the per-user-login-session message
bus.
 Add a session-local.conf and edit that rather than changing this 
 file directly. --

!DOCTYPE busconfig PUBLIC -//freedesktop//DTD D-Bus Bus Configuration
1.0//EN
 http://www.freedesktop.org/standards/dbus/1.0/busconfig.dtd;
busconfig
  !-- Our well-known bus type, don't change this --
  typesession/type

  !-- If we fork, keep the user's original umask to avoid affecting
   the behavior of child processes. --
  keep_umask/

  listenunix:tmpdir=/tmp/listen

  standard_session_servicedirs /

  policy context=default
!-- Allow everything to be sent --
allow send_destination=* eavesdrop=true/
[172.16.48.3]
!-- Allow everything to be received --
allow eavesdrop=true/
!-- Allow anyone to own anything --
allow own=*/
  /policy

  !-- Config files are placed here that among other things, 
   further restrict the above policy for specific services. --
  includedirsession.d/includedir

  servicedir/opt/evo/share/dbus-1/services//servicedir

  !-- This is included last so local configuration can override what's 
   in this standard file --
  include ignore_missing=yessession-local.conf/include

  include if_selinux_enabled=yes
selinux_root_relative=yescontexts/dbus_contexts/include

  !-- For the session bus, override the default relatively-low limits 
   with essentially infinite limits, since the bus is just running 
   as the user anyway, using up bus resources is not something we
need 
   to worry about. In some cases, we do set the limits lower than 
   all available memory if exceeding the limit is almost certainly
a bug, 
   having the bus enforce a limit is nicer than a huge memory leak.
But the 
   intent is that these limits should never be hit. --

  !-- the memory limits are 1G instead of say 4G because they can't
exceed 32-bit signed int max --
  limit name=max_incoming_bytes10/limit
  limit name=max_incoming_unix_fds25000/limit
  limit name=max_outgoing_bytes10/limit
  limit name=max_outgoing_unix_fds25000/limit
  limit name=max_message_size10/limit
  limit name=max_message_unix_fds4096/limit
  limit name=service_start_timeout12/limit  
  limit name=auth_timeout24/limit
  limit name=max_completed_connections10/limit  
  limit name=max_incomplete_connections1/limit
  limit name=max_connections_per_user10/limit
  limit name=max_pending_service_starts1/limit
  limit name=max_names_per_connection5/limit
  limit name=max_match_rules_per_connection5/limit
  limit name=max_replies_per_connection5/limit

/busconfig
___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-04 Thread Paul Smith
Hey Reid; what did you have to do to get this working?  I tried
modifying my configuration then sending HUP to both the system
dbus-daemon and my local dbus-daemon, but when I restart evo I still
don't see any extra factory applications start.

Did you have to kill them outright?  Do they restart?  Did you just log
out/back in?  Reboot?


I don't know why the e-d-s-2.28 stuff would start if you're running off
of the master branch, since those (IIRC) are bonobo services and we
shouldn't be using bonobo anymore?

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-04 Thread Reid Thompson
On Fri, 2009-12-04 at 09:26 -0500, Paul Smith wrote: 
 Hey Reid; what did you have to do to get this working?  I tried
 modifying my configuration then sending HUP to both the system
 dbus-daemon and my local dbus-daemon, but when I restart evo I still
 don't see any extra factory applications start.
 
 Did you have to kill them outright?  Do they restart?  Did you just
 log
 out/back in?  Reboot?
I ended up rebooting ( restarting dbus basically killed my session, so I
just rebooted to 'get clean' ).
Then, I believe, but am not sure, that you actually have to select the
Calendar and Contact view to get dbus to start the backend
 
 
 I don't know why the e-d-s-2.28 stuff would start if you're running
 off
 of the master branch, since those (IIRC) are bonobo services and we
 shouldn't be using bonobo anymore?
 
I wonder if it's getting kicked off by mail-notification..??? Will have
to investigate that...


___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-12-03 Thread Jörgen Scheibengruber
Am Montag, den 30.11.2009, 22:24 +0100 schrieb ext Ross Burton:
 On Mon, 2009-11-30 at 15:17 -0500, Paul Smith wrote:
  PS. I was struck by running ps -aef | grep evo and seeing _NOTHING_
  except the actual evolution binary there... bizarre!
 
 The EDS daemons are now called e-addressbook-factory and
 e-calendar-factory. :)

Something that would IMHO make sense a lot is running each
backend-plugin for those in their own process. Maybe even each
plugin/ESource combination, although that might be something that could
be left to plugin itself to decide and might be hard to do without
breaking api.

But maybe that is planned anyways...

Br,
Jörgen

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-11-30 Thread Ross Burton
On Mon, 2009-11-30 at 14:42 -0500, Paul Smith wrote:
 Details: I'm running on Ubuntu 9.10 64bit.  I am building from the
 latest git master branch, updated as of this morning, for the
 following
 Gnome packages:
 evolution  evolution-data-server  evolution-exchange
 evolution-mapi  evolution-webcal  gtkhtml  libgweather
 libsoup
 libxml2 

Considering that e-d-s master has just been ported to DBus, and
evolution has just had Bonobo removed, I really recommend that you run
the gnome-2-28 branches of the GNOME modules.  Running master means you
acknowledge that stuff may well be broken, and the Evolution modules are
known to be broken (and being fixed) in lots of interesting ways.

Ross
-- 
Ross Burton mail: r...@burtonini.com
  jabber: r...@burtonini.com
   www: http://burtonini.com


signature.asc
Description: This is a digitally signed message part
___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] MAPI support not even close... ?!?! Can I help?

2009-11-30 Thread Paul Smith
On Mon, 2009-11-30 at 19:59 +, Ross Burton wrote:
 Considering that e-d-s master has just been ported to DBus, and
 evolution has just had Bonobo removed, I really recommend that you run
 the gnome-2-28 branches of the GNOME modules.  Running master means
 you acknowledge that stuff may well be broken, and the Evolution
 modules are known to be broken (and being fixed) in lots of
 interesting ways. 

That may be true, but (a) all my IMAP accounts are (so far) working OK,
and (b) I had essentially identical behavior in my MAPI account when I
built the latest gnome-2.28 branches of these packages (I didn't have
the missing subject problem but all the rest of the issues were the
same, and I think the missing subject problem is due to some kind of
glitch in the initial download of the folder data).

So, I don't think the MAPI problems can be laid at the feet of the
current churn in master.

I chose master because, first, the fixes to the critical bugs that
allowed both my entire Exchange Inbox AND my entire Exchange Contacts
list to be deleted from the server by Evo without so much as a warning
were checked into the master branch first, and it was a week or more
with no sign of them being checked into the 2.28 branch.  I think they
have been now.

And second, because I figured developers would be happier about trying
to fix issues on the current master branch rather than do bugfixing and
development on the older branch.

I want MAPI support to actually _work_ in Gnome 2.30.  If that means
running bleeding edge code I'm willing to do that.  If developers prefer
that I test on gnome-2.28 to avoid the overlapping change hassle that's
fine too.  Whatever helps most.


PS. I was struck by running ps -aef | grep evo and seeing _NOTHING_
except the actual evolution binary there... bizarre!

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers