@r5l259, you wrote you consider the Radicale server project to be slowly dying. May I draw your attention to the fact that Radicale recently released an update to new version 3.0 with a rather big set of changes? Maybe what you have interpreted as being neglected has been just lots of work going on silently? 😉
The v3.0.3 got released little more than a week ago, looks to me like this is moving again…

@executive Of course this would be desirable. I have created a feature request for that: https://github.com/ical4j/ical4j/issues/350

HTTP 204 No Content must not have a body and thus Content-Length (which is the definition of 204). It seems like your server has returned HTTP 204 with a body, which does not make sense and can’t be interpreted. Please fix your Web server/application server.

Thank you for the quick response! I read the two issues you linked and yes, it seems my problem is identical to those. Sorry for double posting and not finding those myself :/.

For me, that’s one more reason to leave Horde behind, this software is causing numerous issues in our setup.
What I still don’t really get is why everything seems to be fine as long as I only create/edit tasks from within OpenTasks. Right now I am using a collection with a task that has no value assigned to the “Priority” property (checked by downloading the ics file) and OpenTasks/DAVDroid is syncing the collection without issues.

So for everyone who can’t use the workaround described in https://forums.bitfire.at/topic/1738/task-sync-priority-without-value, having a separate “Android-only” collection seems to help, too.

Thank you very much for the answer!

Thanks for the update and the link!

@andrew said in Events shifted 1 hour with cPanel and Hong Kong Time:

@rfc2822 thanks for the troubleshooting.

I will report to cPanel, possibly via my hosting service provider, and keep everyone posted on the progress.

The following is the update from the hosting service provider.
Dear customer,

According to cPanel reply, the issues was relate to Horde:
https://bugs.horde.org/ticket/13866

Once Horde have resolve the issues, cPanel will including the fix in next cPanel update.

You may consider to add the event at webmail for now.

Thank you.

OK. Your efforts are appreciated. Thanks. ☺

I’ve fixed the problem.

The problem was caused by old calendar entries.
Clearing the howl calendar made the error disappeared.
So Thread could be closed.

Thx for your answer!

DAvdroid is not a calendar app, but a sync adapter. You can use the calendar app of your choice. All devices should have a calendar app pre-installed.

ok, issue solved!

It was a problem with Horde. I shared a calendar with another user who is on another server. Every time I clicked the calendar in the web interface, I was logged out. After a while the link to the problem described here came into my mind. Now problem is fixed.

Hi,
sorry, but I do not know how to reproduce that - otherwise I would do my best to avoid it. Just arbitrary it happens sometimes. Maybe it is caused by a bad connection so that the sync cannot be finished.

Setup is:
Server: Horde 5.2.13 with Kronolith H5 (4.2.19)
logfile is difficult cause I cannot provoke the problem

But … on the other hand: would it be too complicated to add a button to check for dupes? Hopefully someone else comes around the corner who has the same problem at times. The solution of the problem is rather complicated, having to resync the whole calender, not to mention the loss of entries older than the set time (that are not synced any more but stay on the device nevertheless)

regards

When uploading an event at 16 Feb 2017 10:00 Eastern standard time, DAVdroid puts this event + ical4j timezone definition to the server:

BEGIN:VCALENDAR VERSION:2.0 PRODID:+//IDN bitfire.at//DAVdroid/1.4.0.2 ical4android ical4j/2.x BEGIN:VEVENT DTSTAMP:20170219T175556Z UID:20170219T175554Z-b17c600e48506c59 DTSTART;TZID=America/New_York:20170216T100000 DTEND;TZID=America/New_York:20170216T110000 SUMMARY:Test STATUS:TENTATIVE END:VEVENT BEGIN:VTIMEZONE TZID:America/New_York TZURL:http://tzurl.org/zoneinfo/America/New_York X-LIC-LOCATION:America/New_York BEGIN:DAYLIGHT TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:EDT DTSTART:20070311T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:EST DTSTART:20071104T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU END:STANDARD END:VTIMEZONE END:VCALENDAR

(empty lines for visual separation)

Then, Horde converts this to this and sends it back (as an updated event):

BEGIN:VCALENDAR VERSION:2.0 X-WR-CALNAME:Cal Of Thrones PRODID:-//The Horde Project//Horde iCalendar Library//EN BEGIN:VEVENT DTSTART;TZID=America/New_York:20170216T090000 DTEND;TZID=America/New_York:20170216T100000 DTSTAMP:20170219T175649Z UID:20170219T175554Z-b17c600e48506c59 CREATED:20170219T175634Z LAST-MODIFIED:20170219T175634Z SUMMARY:Test CLASS:PUBLIC STATUS:TENTATIVE TRANSP:OPAQUE END:VEVENT BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT TZOFFSETFROM:-0400 !!!! TZOFFSETTO:-0400 DTSTART:20070311T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYMONTHDAY=8,9,10,11,12,13,14;BYDAY=1SU TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0400 TZOFFSETTO:-0500 DTSTART:20071104T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU TZNAME:EST END:STANDARD END:VTIMEZONE END:VCALENDAR

As you can see, the daylight definition switches from a -4:00 offset to -4:00, while standard time switches from -4 to -5.

So this is a timezone problem of Horde (specifically in the DAYLIGHT definition) and not a DAVdroid problem. Please contact Horde regarding this.

I don’t know why SOGo has this problem, too. Maybe there’s a problem with the tzdata, but in this case, I can’t do anything, too.

If there are any indications that this could be a DAVdroid problem, feel free to post here again.

Can you send the test account to play@bitfire.at (OpenPGP)?

Okay solved the problem myself. It may be useful for someone else.

I am a bit paranoid so my server have a strict firewall blocking anything that has not been white listed. Even on output.
Horde need to synchronize the timezone using caldav. There for it tries to get timezone infos using a ftp:
$conf[‘timezone’][‘location’] =‘ftp://ftp.iana.org/tz/tzdata-latest.tar.gz’;
It was blocked by my firewall and therefore blocking sync.

thanks for the app

Upgrading to Horde Framework 5.2.14 solved the problem.
Thanks for the tip.

Apparently A fatal error has occurred Bind failed: Server is unwilling to perform is horde/LDAP’s way to tell me the password is wrong… great. So I re-entered the password that I have not changed in DAVdroid and it works now. Hope this helps anyone else who is lost like me.

The NullPointerException is just informational for the “Received invalid iCalendar, ignoring” line, so everything except this event should have been synchronized.

ical4j throws this exception when trying to parse the VTIMEZONE because there’s an invalid component in your VTIMEZONE definition:

BEGIN:DAYLIGHT TZOFFSETFROM:+0119 TZOFFSETTO:+0119 TZNAME:NST END:DAYLIGHT

Time zone components must have a DTSTART value, otherwise they’re useless and invalid. This is similar to DAVdroid doesn’t understand America/Montevideo VTIMEZONE provided by Horde and Can’t see calendar events.

Because the time zone is not parseable, ical4j and thus ical4android/DAVdroid cannot process the event. To fix it, please use a correct time zone definition.

I also suggest reporting this to Horde.

This horde problem was already seen in beta tests, see Address books duplicated by -system- group (I have moved the topic to the public forum).