5:23 p.m.
Hi,
I switched to the new syncevo-http-server script from git, and now my
syncevolution-dbus-server terminates every 10 minutes. Any hints what
is causing this?
Regards,
Tino
6:43 p.m.
On Mi, 2011-02-02 at 06:23 +0000, Tino Keitel wrote:
I switched to the new syncevo-http-server script from git, and now
my
syncevolution-dbus-server terminates every 10 minutes. Any hints what
is causing this?
That's the normal behavior of syncevo-dbus-server. It's shutting down
when not in use. It should be restarted by syncevo-http-server once
there is an incoming connection (if you use the --start-syncevolution
option) or the D-Bus daemon (if you have it installed such that the
daemon can find the service description).
Is there a reason why you want it to keep running, or does the restart
mechanism fail?
The 10 minute delay can be increased via the -d command line option, but
there's currently no way to set that when starting via
syncevo-http-server, except by patching the script.
--
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.
7 p.m.
On Wed, Feb 02, 2011 at 08:43:53 +0100, Patrick Ohly wrote:
On Mi, 2011-02-02 at 06:23 +0000, Tino Keitel wrote:
> I switched to the new syncevo-http-server script from git, and now my
> syncevolution-dbus-server terminates every 10 minutes. Any hints what
> is causing this?
That's the normal behavior of syncevo-dbus-server. It's shutting down
when not in use. It should be restarted by syncevo-http-server once
there is an incoming connection (if you use the --start-syncevolution
option) or the D-Bus daemon (if you have it installed such that the
daemon can find the service description).
Is there a reason why you want it to keep running, or does the restart
mechanism fail?
I used to use 2 services on my syncevolution server: one that sets up a
dbus session bus and starts syncevo-dbus-server, and one service for
the syncevo-http-server script. The dbus-daemon --session is killed
when the syncevo-dbus-server service is stopped, so that no stale
processes exist.
I guess I should just use syncevo-http-server --start-dbus-session and
trust the script to not leave stale dbus-server --session processes.
When I do this, I get this:
syncevo-http: D-Bus session found (DISPLAY=
DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-gRbqTYtIbL,guid=eb2b4568aa8af99d0b0f083d4d490c68),
but starting a new D-Bus session as requested
I don't know where it gets the old session. It doesn't seem to exist,
and the service should be started in a clean environment.
Regards,
Tino
7:25 p.m.
On Mi, 2011-02-02 at 08:00 +0000, Tino Keitel wrote:
I guess I should just use syncevo-http-server --start-dbus-session
and
trust the script to not leave stale dbus-server --session processes.
Yes, please give it a try and report back if it doesn't work.
When I do this, I get this:
syncevo-http: D-Bus session found (DISPLAY=
DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-gRbqTYtIbL,guid=eb2b4568aa8af99d0b0f083d4d490c68),
but starting a new D-Bus session as requested
I don't know where it gets the old session. It doesn't seem to exist,
and the service should be started in a clean environment.
Are you logged into the machine with a running X session?
In that case the script finds the X session and from there the D-Bus
session. This catches the situation that someone uses the script from
cron on his desktop, because then he would normally end up with
Evolution Data Server running twice: once inside the cron session, once
inside the normal X session.
--
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.
7:27 p.m.
On Wed, Feb 02, 2011 at 09:25:47 +0100, Patrick Ohly wrote:
On Mi, 2011-02-02 at 08:00 +0000, Tino Keitel wrote:
> I guess I should just use syncevo-http-server --start-dbus-session and
> trust the script to not leave stale dbus-server --session processes.
Yes, please give it a try and report back if it doesn't work.
> When I do this, I get this:
>
> syncevo-http: D-Bus session found (DISPLAY=
>
DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-gRbqTYtIbL,guid=eb2b4568aa8af99d0b0f083d4d490c68),
> but starting a new D-Bus session as requested
>
> I don't know where it gets the old session. It doesn't seem to exist,
> and the service should be started in a clean environment.
Are you logged into the machine with a running X session?
In that case the script finds the X session and from there the D-Bus
session. This catches the situation that someone uses the script from
cron on his desktop, because then he would normally end up with
Evolution Data Server running twice: once inside the cron session, once
inside the normal X session.
It is a headless server and I'm logged in via SSH. A "ps aux | grep
dbus-daemon" shows only the --system daemon.
Regards,
Tino
7:54 p.m.
On Mi, 2011-02-02 at 08:27 +0000, Tino Keitel wrote:
On Wed, Feb 02, 2011 at 09:25:47 +0100, Patrick Ohly wrote:
> On Mi, 2011-02-02 at 08:00 +0000, Tino Keitel wrote:
> > I guess I should just use syncevo-http-server --start-dbus-session and
> > trust the script to not leave stale dbus-server --session processes.
>
> Yes, please give it a try and report back if it doesn't work.
>
> > When I do this, I get this:
> >
> > syncevo-http: D-Bus session found (DISPLAY=
> >
DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-gRbqTYtIbL,guid=eb2b4568aa8af99d0b0f083d4d490c68),
> > but starting a new D-Bus session as requested
> >
> > I don't know where it gets the old session. It doesn't seem to exist,
> > and the service should be started in a clean environment.
>
> Are you logged into the machine with a running X session?
>
> In that case the script finds the X session and from there the D-Bus
> session. This catches the situation that someone uses the script from
> cron on his desktop, because then he would normally end up with
> Evolution Data Server running twice: once inside the cron session, once
> inside the normal X session.
It is a headless server and I'm logged in via SSH. A "ps aux | grep
dbus-daemon" shows only the --system daemon.
Is the DBUS_SESSION_BUS_ADDRESS env variable perhaps set anyway, despite
not really having that daemon running? After looking at the code that's
pretty much what the message above says. The logic I mentioned (look up
sessions, find X session) is elsewhere.
--
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.
8 p.m.
On Wed, Feb 02, 2011 at 09:54:58 +0100, Patrick Ohly wrote:
[...]
Is the DBUS_SESSION_BUS_ADDRESS env variable perhaps set anyway,
despite
not really having that daemon running? After looking at the code that's
pretty much what the message above says. The logic I mentioned (look up
sessions, find X session) is elsewhere.
Oops, I run the HTTP server with the wrong environment. The one from
the old syncevo-dbus-server service was provided. I fixed my service
script and now this message is gone. Sorry for the noise.
Regards,
Tino
4246
days inactive
4246
days old
syncevolution@synevolution.org
6 comments
2 participants
participants (2)
-
Patrick Ohly -
Tino Keitel