Hi,

I'm using the x64 version of Rainlendar (always the latest beta),
and every time Windows 7 resumes from Standby, Rainlendar gives an error that the Google calendar cannot be accessed. Also, the weather widget fails to update.

It looks like Rainlendar is trying to do a refresh too fast, when the network connection isn't available.

In the extended options there is a setting for a time period to wait afer a start, but this setting seems to have no effect for a resume after standby.

Michael

The exact error message:

Google Error (Google Calendar)
Unable to download the calendars.
Server returned result: (6).

The rest of the text is localized, meaning there's an error in resolving the host. Make sure your network connection is working and no firewall is blocking the connection.

Loom wrote:
The exact error message:

Google Error (Google Calendar)
Unable to download the calendars.
Server returned result: (6).

The rest of the text is localized, meaning there's an error in resolving the host. Make sure your network connection is working and no firewall is blocking the connection.
Did you try ticking "Keep offline copy" for those calendars? That way Rainlendar will read the stored version and will try to update as soon as connection is actually available.
OR you can try turning "Supress Errors" as well.

Yes, I'm using an offline copy for the Google calendar all the time.

I could, however, surpress the error messages. But I think fixing this bug would be more helpful.

By the way, I set the refresh rate for the calendar to 60 minutes.

So, when I enter standby, and wake it up immediately, there is no error.
But when the machine wakes up from a longer standby (longer then those 60 minutes), it looks like Rainlendar recognizes that it's time for a refresh, and tries do to so immediately, when the network connection isn't yet ready.

That is true. The problem is that the thread which polls the events from the server gets started before the application receives the notification that computer was resumed from suspend. I'll see if there is something that can be done about this.

Maybe inserting an extra 5 seconds of delay before letting the thread poll the updates ?