Re: [TOOLS-DEVELOPMENT] ical server issue

[Lou Berger <[e-mail]>]

Well in trying a test with removing Z I found something really interesting:

All works well if the non-https version is used
(http://ical.ietf.org/public.php/IAOC/calendar)

Is there perhaps something odd on the server's time setting (assuming
google is somehow using tls time...)?

Lou


On 9/14/2015 12:15 PM, Glen wrote:
> Hi Lou -
>
> I"m trying to follow this... but in the thread you link to, and the
> 4/1 response you cite, it seems that Google is talking about the
> manual import of ICS files, and saying that it is fixed, which it is. 
> I can manually import the ICS file, and the timezones are all correct
> on my Google Calendar.   It seems that the problem only occurs with
> calendar subscriptions.
>
> In addition, you mention someone removing UTC times from the ICS... 
> When I do a survey of the exported ICS, it seems that about half the
> times are in UTC format (e.g. DTSTART:20131219T150000Z), and the other
> roughly half appear to have time zone specifiers.  (e.g. 
> DTSTART;TZID=America/New_York:20150910T120000).  I do not have any way
> of automatically making the calendar server remove all the "Z" times
> from its feed, and converting them to some time of local time - that
> is a function of each calendar item's creation and time zone settings.
>
> Interestingly, the item in question - the one malfunctioning - is the
> one I cited above, and it is *not* a "Z" time, but is in fact a
> localized timezone time.  I reviewed the file output, and compared
> them to 2445/5545, and the output seems to be exactly compliant.  We
> also have a properly-constructed VTIMEZONE stanza for it, and it is,
> as already cited, working for all other platforms (including Google's
> manual import.)
>
> So I'm still in the same place - our server is functioning, and
> functioning correctly, and RFC-compliant, and although I can confirm
> that Google is mangling this on import, I do not have any connections
> to Google that would enable me to "run this down..." much as I, too,
> would like to see this solved for their platform.
>
> I am open to any thoughts or suggestions... Do we know anyone at
> Google who participates in the IETF that might have some access to
> work on this, for example?
>
> Glen
>
>
> On Sun, Sep 13, 2015 at 8:29 AM, Lou Berger <[e-mail]
> <mailto:[e-mail]>> wrote:
>
>     Glenn,
>
>     Looks like others have seen similar issues and worked with Google
>     to resolve, see links below. Given that there will likely be other
>     Google users of the ietf ical server it's probably worth running
>     this to ground...
>
>     Lou
>
>     -
>     https://productforums.google.com/forum/m/#!topic/calendar/GouovR3g4LY
>     <https://productforums.google.com/forum/m/#%21topic/calendar/GouovR3g4LY>.
>     Look at the 4/1 response.
>
>     - I lost the link, but someone solve the issues by removing any Z
>     (utc) based times from the ics
>
>
>
>
>     On September 11, 2015 2:00:09 PM Lou Berger <[e-mail]
>     <mailto:[e-mail]>> wrote:
>
>         another data point:
>         https://ical.ietf.org/public.php/IAOC/finance
>
>         works fine...
>
>         On 9/11/2015 1:00 PM, Glen wrote:
>
>             Here's an interesting data point:
>
>             If I download the file at the
>             https://ical.ietf.org/public.php/IAOC/calendar URL
>             manually, and save
>             it as an ICS file on my desktop, and then I "Import" that
>             file into
>             Google, the times display correctly.
>
>             It is only if I ask Google to "subscribe" to the URL in
>             live-feed mode
>             that it seems to incorrectly compute the times.
>
>
>
>
>             As there is no difference in the data flow or what is
>             pulled (since
>             the Calendar server sends everything via HTTP in the same
>             way), I
>             wonder if something in Google's subscription processor is
>             tripping on
>             something somewhere.
>
>             That's not a solution - because you obviously want the
>             live feed - but
>             it does provide an additional point of validation, if you
>             like, of the
>             data.
>
>             Glen
>
>
>
>
>
>
>
>             On Fri, Sep 11, 2015 at 9:52 AM, Glen <[e-mail]
>             <mailto:[e-mail]>
>             <mailto:[e-mail] <mailto:[e-mail]>>> wrote:
>
>                 Hi Lou -
>
>                 AMS ([e-mail]
>             <mailto:[e-mail]> <mailto:[e-mail]
>             <mailto:[e-mail]>>) would be
>                 the correct place to report a problem with the
>             calendar *server* -
>                 for example, if the calendar server were down, or you
>             couldn't get
>                 access, or something like that.
>
>                 But I confess I have no idea how or where to "report"
>             the problem
>                 you're encountering.
>
>                 When I view the feed you mention, I see the meeting
>             coded as:
>
>                 DTEND;TZID=America/New_York:20150910T130000
>                 DTSTART;TZID=America/New_York:20150910T120000
>
>                 So the calendar server appears to be delivering it
>             correctly, as
>                 you point out.
>
>                 The question seems to be why Google calendar is
>             misinterpreting
>                 the coding here.
>
>                 I also use Gogole calendar, and it is also
>             misinterpreting the
>                 coding for me.... but I cannot tell why, nor would I
>             know to whom
>                 that should be reported.
>
>                 I welcome thoughts on this matter!
>
>                 Glen
>                 Glen Barney
>                 IT Director
>                 AMS (IETF Secretariat)
>
>
>
>                 On Fri, Sep 11, 2015 at 7:52 AM, Lou Berger
>             <[e-mail] <mailto:[e-mail]>
>                 <mailto:[e-mail] <mailto:[e-mail]>>>
>             wrote:
>
>                     So what's the right way to report an issue with
>             ical.ietf.org <http://ical.ietf.org>
>                     <http://ical.ietf.org>?
>
>                     I see an issue with a meeting from yesterday
>             9/10/15 on
>                     https://ical.ietf.org/public.php/IAOC/calendar and
>             it looks
>                     like the
>                     issues is also there for future meetings.
>
>                     The meeting correctly shows in Thunderbird as
>             occurring at
>                     noon ET, but
>                     on google calendar it shows at 10am ET. 
>             Interesting, this has
>                     not been
>                     an issue for other calendars, e.g., finance.
>
>                     Thanks,
>                     Lou
>
>
>
>                     _______________________________________________
>                     TOOLS-DEVELOPMENT mailing list
>                     [e-mail]
>             <mailto:[e-mail]>
>             <mailto:[e-mail]
>             <mailto:[e-mail]>>
>                    
>             https://www.ietf.org/mailman/listinfo/tools-development
>
>
>
>
>
>         _______________________________________________
>         TOOLS-DEVELOPMENT mailing list
>         [e-mail] <mailto:[e-mail]>
>         https://www.ietf.org/mailman/listinfo/tools-development
>
>
>
>