4:17 a.m.
On Di, 2010-03-16 at 16:34 +0000, Stefano Zacchiroli wrote:
On Tue, Mar 16, 2010 at 05:03:40PM +0100, Patrick Ohly wrote:
> Congwu has access to a N900, but as he wasn't able to reproduce the
> problem, we need your logs. It might also be specific to your
> data. And no, Nokia is not sending around hardware just because some
> developer asks for it.
Well, I got one N900 as a distro dev, that made me believe Nokia was
interested in handing out phones to people doing development on the
device.
I got a 770 and N810 the same way, but that was when I worked on this in
my spare time. I didn't ask to get included in the N900 program and so
far Nokia has not offered free devices again, at least as far as I know.
FWIW, in the thread(s) I got the impression that none of the
devs actually had a N900 and/or tried to reproduce the issue.
You are right, that Congwu has one was probably only mentioned in one of
the earlier threads about the N900.
On the same vein, I don't have any particularly sensitive date in
my
calendar. If you can guarantee that it won't go public for download, I
can and hover the data to syncevolution devs too. Would having the data
enable some more in-depth debugging?
Not at our end. If it turns out to be related to the specific *and* we
can get a Nokia developer to investigate this, then we might come back
to your offer.
> It uses a single .ics file per calendar. In your mail with the
log you
> mentioned that events were only duplicated inside that calendar, so I'll
> ignore that part of the problem unless it happens again.
In fact, I reported that some events got duplicated in the same calendar
(i.e. two entries corresponding to the same event in the same calendar)
That I understand. We send events from calendar A to the device, get
some of the same events back when syncing A again, so we duplicate them
in A.
and some others in different calendars (i.e. two entries
corresponding
to the same event in two different calendars).
That's what I don't understand. When configured to use calendar A,
SyncEvolution should never touch a different calendar B.
> The log confirms that the SyncML client in the N900 is confused,
as I
> suspected. It explicitly adds all of these duplicates to the host in a
> normal two-way sync. Probably the change tracking on the device is
> broken, perhaps because of the amount of calendar items that you have.
OK, thanks for this investigation. What is the "normal" amount of events
and/or time period which encompasses all the events? I don't delete past
event just out of laziness, I don't particularly mind deleting old
entries. Before doing so however, it might be useful to have some
figures to check at which point the problem starts to appear.
Congwu might be able to test this with our test tool so that you don't
have to test this manually. A few thousand events is a bit at the high
end, but I would expect it to work.
--
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.