From bremner at unb.ca Tue Mar 8 14:25:27 2011 From: bremner at unb.ca (David Bremner) Date: Tue, 08 Mar 2011 10:25:27 -0400 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> Message-ID: <87bp1lbsfc.fsf@zancas.localnet> On Fri, 25 Feb 2011 21:00:30 +0100, Patrick Ohly wrote: > > No, they need to be rewritten. Matthew confirmed that this happens > outside of the backends, so I suspect that it is necessary to run the > Evolution UI to trigger the rewriting of gconf keys. > > I can confirm that running the UI rewrites the keys for me. Tino, does > that sound plausible to you? Does it fix your problem? > I seem to be encountering a similar problem: I am using evolution 2.32.2-1 from debian sid, and running syncevolution causes ,---- | (e-addressbook-factory:24865): e-data-server-WARNING **: e_source_get_uri () called on source with no absolute URI! | zsh: segmentation fault /usr/lib/evolution/e-addressbook-factory `---- I checked in gconf-editor and all of the keys Matthew mentioned have base_uri="local:" This is with syncevolution 1.1.99.3 from git. I have previously migrated syncevolution to the new format (with 1.1.99.2, I think) , and it seemed to be working OK, at least in a few simple tests. Since then, I guess several libraries have been upgraded on this machine. I'll include the (long) list of evolution dependencies here, in case the versions strike someone as suspicious. Versions of packages evolution depends on: ii dbus 1.2.24-4 simple interprocess messaging syst ii debconf [debconf-2.0] 1.5.36.1 Debian configuration management sy ii evolution-common 2.32.2-1 architecture independent files for ii evolution-data-server 2.32.2-2 evolution database backend server ii gconf2 2.28.1-6 GNOME configuration database syste ii gnome-icon-theme 2.30.3-2 GNOME Desktop icon theme ii libatk1.0-0 1.30.0-1 The ATK accessibility toolkit ii libc6 2.11.2-10 Embedded GNU C Library: Shared lib ii libcairo2 1.8.10-6 The Cairo 2D vector graphics libra ii libcamel1.2-19 2.32.2-2 The Evolution MIME message handlin ii libcanberra-gtk0 0.24-1 Gtk+ helper for playing widget eve ii libcanberra0 0.24-1 a simple abstract interface for pl ii libdbus-1-3 1.2.24-4 simple interprocess messaging syst ii libdbus-glib-1-2 0.88-2.1 simple interprocess messaging syst ii libebackend1.2-0 2.32.2-2 Utility library for evolution data ii libebook1.2-10 2.32.2-2 Client library for evolution addre ii libecal1.2-8 2.32.2-2 Client library for evolution calen ii libedataserver1.2-14 2.32.2-2 Utility library for evolution data ii libedataserverui1.2-11 2.32.2-2 GUI utility library for evolution ii libegroupwise1.2-13 2.32.2-2 Client library for accessing group ii libenchant1c2a 1.6.0-1 a wrapper library for various spel ii libevolution 2.32.2-1 evolution libraries ii libfontconfig1 2.8.0-2.1 generic font configuration library ii libfreetype6 2.4.2-2.1 FreeType 2 font engine, shared lib ii libgail18 2.20.1-2 GNOME Accessibility Implementation ii libgconf2-4 2.28.1-6 GNOME configuration database syste ii libgdata7 0.6.4-3 Library for accessing GData webser ii libglib2.0-0 2.28.1-1+b1 The GLib library of C routines ii libgnome-desktop-2-17 2.30.2-2 Utility library for loading .deskt ii libgtk2.0-0 2.20.1-2 The GTK+ graphical user interface ii libgtkhtml-editor-3.14 3.32.2-1 HTML rendering/editing library - e ii libgtkhtml3.14-19 3.32.2-1 HTML rendering/editing library - r ii libgweather1 2.30.3-1 GWeather shared library ii libical0 0.44-3 iCalendar library implementation i ii libice6 2:1.0.6-2 X11 Inter-Client Exchange library ii libnotify1 [libnotify1 0.5.0-2 sends desktop notifications to a n ii libnspr4-0d 4.8.6-1 NetScape Portable Runtime Library ii libnss3-1d 3.12.8-1 Network Security Service libraries ii libpango1.0-0 1.28.3-1+squeeze2 Layout and rendering of internatio ii libsm6 2:1.1.1-1 X11 Session Management library ii libsoup2.4-1 2.30.2-1 an HTTP library implementation in ii libsqlite3-0 3.7.4-1 SQLite 3 shared library ii libstartup-notificatio 0.10-1 library for program launch feedbac ii libunique-1.0-0 1.1.6-2 Library for writing single instanc ii libxml2 2.7.8.dfsg-2 GNOME XML library ii zlib1g 1:1.2.3.4.dfsg-3 compression library - runtime From patrick.ohly at gmx.de Tue Mar 8 16:06:45 2011 From: patrick.ohly at gmx.de (Patrick Ohly) Date: Tue, 08 Mar 2011 08:06:45 -0800 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <87bp1lbsfc.fsf@zancas.localnet> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> Message-ID: <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> On Di, 2011-03-08 at 14:25 +0000, David Bremner wrote: > On Fri, 25 Feb 2011 21:00:30 +0100, Patrick Ohly wrote: > > > > No, they need to be rewritten. Matthew confirmed that this happens > > outside of the backends, so I suspect that it is necessary to run the > > Evolution UI to trigger the rewriting of gconf keys. > > > > I can confirm that running the UI rewrites the keys for me. Tino, does > > that sound plausible to you? Does it fix your problem? > > > > I seem to be encountering a similar problem: I am using evolution > 2.32.2-1 from debian sid, and running syncevolution causes > > ,---- > | (e-addressbook-factory:24865): e-data-server-WARNING **: e_source_get_uri () called on source with no absolute URI! > | zsh: segmentation fault /usr/lib/evolution/e-addressbook-factory > `---- > > I checked in gconf-editor and all of the keys Matthew mentioned have base_uri="local:" > > This is with syncevolution 1.1.99.3 from git. I have previously migrated > syncevolution to the new format (with 1.1.99.2, I think) , and it seemed > to be working OK, at least in a few simple tests. Since then, I guess > several libraries have been upgraded on this machine. Did you restart syncevo-dbus-server after rewriting the gconf keys? I doubt that this is the reason (their values shouldn't be cached), but I want to rule out this explanation. What does "syncevolution" (with no args) say about the databases and their URIs? Finally, just for debugging purposes, what URI is passed to e-addressbook-factory over D-Bus? Run dbus-monitor and watch out for the method which opens the address book (not exactly sure what it is). -- Best Regards, Patrick Ohly The content of this message is my personal opinion only and although I am an employee of Intel, the statements I make here in no way represent Intel's position on the issue, nor am I authorized to speak on behalf of Intel on this matter. From bremner at unb.ca Tue Mar 8 16:55:28 2011 From: bremner at unb.ca (David Bremner) Date: Tue, 08 Mar 2011 12:55:28 -0400 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> Message-ID: <87zkp5a6wv.fsf@zancas.localnet> On Tue, 08 Mar 2011 08:06:45 -0800, Patrick Ohly wrote: > > Did you restart syncevo-dbus-server after rewriting the gconf keys? I > doubt that this is the reason (their values shouldn't be cached), but I > want to rule out this explanation. Yes. By co-incidence, I even rebooted the machine ;). > > What does "syncevolution" (with no args) say about the databases and > their URIs? Evolution Address Book = Evolution Contacts = evolution-contacts: Personal (local:/system) Evolution Calendar = evolution-calendar: Personal (local:/system) Birthdays & Anniversaries (contacts:///) Evolution Task List = Evolution Tasks = evolution-tasks: Personal (local:/system) Evolution Memos = evolution-memos: Personal (local:/system) > > Finally, just for debugging purposes, what URI is passed to > e-addressbook-factory over D-Bus? Run dbus-monitor and watch out for the > method which opens the address book (not exactly sure what it is). I think this is the relevant call: ,---- | method call sender=:1.28 -> dest=:1.34 serial=19 path=/org/gnome/evolution/dataserver/addressbook/BookFactory; interface=org.gnome.evolution.dataserver.addressbook.BookFactory; member=getBook | string " | `---- I don't really speak dbus, but after that comes something that at a wild guess looks like e-addressbook-factory crashing. ,---- | signal sender=org.freedesktop.DBus -> dest=(null destination) serial=9 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=NameOwnerChanged | string "org.gnome.evolution.dataserver.AddressBook" | string ":1.34" | string "" `---- d From mbarnes at redhat.com Tue Mar 8 17:10:44 2011 From: mbarnes at redhat.com (Matthew Barnes) Date: Tue, 08 Mar 2011 12:10:44 -0500 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <87zkp5a6wv.fsf@zancas.localnet> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> <87zkp5a6wv.fsf@zancas.localnet> Message-ID: <1299604248.17777.2.camel@localhost.localdomain> On Tue, 2011-03-08 at 12:55 -0400, David Bremner wrote: > I don't really speak dbus, but after that comes something that at a wild > guess looks like e-addressbook-factory crashing. > > ,---- > | signal sender=org.freedesktop.DBus -> dest=(null destination) serial=9 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=NameOwnerChanged > | string "org.gnome.evolution.dataserver.AddressBook" > | string ":1.34" > | string "" > `---- That would be my guess too. Can you try running e-addressbook-factory from gdb and see if you can capture a stack trace of the crash? Here's a "how-to" if you need it: http://live.gnome.org/GettingTraces/Details Matthew Barnes From bremner at unb.ca Tue Mar 8 20:56:36 2011 From: bremner at unb.ca (David Bremner) Date: Tue, 08 Mar 2011 16:56:36 -0400 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <1299604248.17777.2.camel@localhost.localdomain> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> <87zkp5a6wv.fsf@zancas.localnet> <1299604248.17777.2.camel@localhost.localdomain> Message-ID: <87fwqx9vqz.fsf@zancas.localnet> On Tue, 08 Mar 2011 12:10:44 -0500, Matthew Barnes wrote: > On Tue, 2011-03-08 at 12:55 -0400, David Bremner wrote: > > I don't really speak dbus, but after that comes something that at a wild > > guess looks like e-addressbook-factory crashing. > > > > ,---- > > | signal sender=org.freedesktop.DBus -> dest=(null destination) serial=9 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=NameOwnerChanged > > | string "org.gnome.evolution.dataserver.AddressBook" > > | string ":1.34" > > | string "" > > `---- > > That would be my guess too. Can you try running e-addressbook-factory > from gdb and see if you can capture a stack trace of the crash? Thanks to a hint from Ken Bloom, I realized the problem was the evolution related library versions. At least libebook-1.2 and libcamel-1.2 were at revisions corresponding to evolution 2.30.3 in the compilation environment. With these updated, the problem goes away. Some preliminary testing suggests that the version of syncevolution compiled with the 2.32.2 evolution libs won't work with evolution 2.30.3. David From patrick.ohly at gmx.de Tue Mar 8 21:33:49 2011 From: patrick.ohly at gmx.de (Patrick Ohly) Date: Tue, 08 Mar 2011 13:33:49 -0800 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <87fwqx9vqz.fsf@zancas.localnet> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> <87zkp5a6wv.fsf@zancas.localnet> <1299604248.17777.2.camel@localhost.localdomain> <87fwqx9vqz.fsf@zancas.localnet> Message-ID: <1299620029.4561.391.camel@pohly-mobl1.ikn.intel.com> On Di, 2011-03-08 at 20:56 +0000, David Bremner wrote: > Some preliminary testing suggests that the version of syncevolution > compiled with the 2.32.2 evolution libs won't work with evolution > 2.30.3. I'm not surprised by that. Evolution 2.30 client libs simply seem not to be compatible with 2.32 EDS, and vice versa. But this shouldn't be a problem for Debian, right? Just ensure that everything in the same repo uses EDS from that same repo. To prevent breakage during upgrades, perhaps EDS 2.32 libs should conflict with older EDS libs? They are installable in parallel, but mixing them does not really lead to a usable system. -- Best Regards, Patrick Ohly The content of this message is my personal opinion only and although I am an employee of Intel, the statements I make here in no way represent Intel's position on the issue, nor am I authorized to speak on behalf of Intel on this matter. From kbloom at gmail.com Tue Mar 8 21:42:54 2011 From: kbloom at gmail.com (Chanoch (Ken) Bloom) Date: Tue, 08 Mar 2011 15:42:54 -0600 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <1299620029.4561.391.camel@pohly-mobl1.ikn.intel.com> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> <87zkp5a6wv.fsf@zancas.localnet> <1299604248.17777.2.camel@localhost.localdomain> <87fwqx9vqz.fsf@zancas.localnet> <1299620029.4561.391.camel@pohly-mobl1.ikn.intel.com> Message-ID: <1299620574.25794.14.camel@cat-in-the-hat.dnsalias.com> On Tue, 2011-03-08 at 13:33 -0800, Patrick Ohly wrote: > On Di, 2011-03-08 at 20:56 +0000, David Bremner wrote: > > Some preliminary testing suggests that the version of syncevolution > > compiled with the 2.32.2 evolution libs won't work with evolution > > 2.30.3. > > I'm not surprised by that. Evolution 2.30 client libs simply seem not to > be compatible with 2.32 EDS, and vice versa. But this shouldn't be a > problem for Debian, right? Just ensure that everything in the same repo > uses EDS from that same repo. I'm not the Debian maintainer for this package, but I doubt that it will be much of a problem. Once they fix the bug that prevents them from building syncevolution (which I see you're already aware of), this should get fixed automatically. > To prevent breakage during upgrades, perhaps EDS 2.32 libs should > conflict with older EDS libs? They are installable in parallel, but > mixing them does not really lead to a usable system. Might be worth filing a bug with Debian about that specifically, but I'm not an expert about this, so I don't want to file the bug myself. Probably the new evolution-data-server should conflict with the old libraries. --Ken From mbarnes at redhat.com Tue Mar 8 23:11:08 2011 From: mbarnes at redhat.com (Matthew Barnes) Date: Tue, 08 Mar 2011 18:11:08 -0500 Subject: [SyncEvolution] Issues with evolution 2.32 and syncevolution 1.1.1a In-Reply-To: <1299620029.4561.391.camel@pohly-mobl1.ikn.intel.com> References: <20110215205804.GA17672@mac.home> <1297865461.5364.505.camel@pohly-mobl1.ikn.intel.com> <1297870847.2609.37.camel@localhost.localdomain> <1297871999.5364.559.camel@pohly-mobl1.ikn.intel.com> <1298664030.3995.456.camel@pohly-mobl1.ikn.intel.com> <87bp1lbsfc.fsf@zancas.localnet> <1299600405.4561.289.camel@pohly-mobl1.ikn.intel.com> <87zkp5a6wv.fsf@zancas.localnet> <1299604248.17777.2.camel@localhost.localdomain> <87fwqx9vqz.fsf@zancas.localnet> <1299620029.4561.391.camel@pohly-mobl1.ikn.intel.com> Message-ID: <1299625870.15517.7.camel@localhost.localdomain> On Tue, 2011-03-08 at 13:33 -0800, Patrick Ohly wrote: > I'm not surprised by that. Evolution 2.30 client libs simply seem not to > be compatible with 2.32 EDS, and vice versa. But this shouldn't be a > problem for Debian, right? Just ensure that everything in the same repo > uses EDS from that same repo. FWIW, I've since added version numbers to our D-Bus bus names for this very reason. Had I thought of it earlier, the address book service in 2.30 would have been named: org.gnome.evolution.dataserver.Addressbook0 and in 2.32 named org.gnome.evolution.dataserver.Addressbook1 to account for the API break that made them incompatible. Then you would have seen a much more informative error message from D-Bus. Hopefully these compatibility problems will be a thing of past for 3.0 and beyond.