Re: [Evolution] Errors receiving mail via POP

[Pete Biggs <pete biggs org uk>]

> Yup.  Fetchmail just sleeps however long the ...rc file requires and
> tries again.  There appears to be a limit to how long fetchmail will
> wait for an appropriate reply from a server.  From the man page:
>
>         The timeout option allows you to set a server-nonresponse
>         timeout in seconds.  If a mailserver does not send  a greeting
>         message or respond to commands for the given number of seconds,
>         fetchmail will drop the connection to it. ... If a given
>         connection  receives  too  many timeouts in succession,
>         fetchmail will consider it wedged and stop retrying.  The
>         calling user will be notified by email if this happens.
>
> This might be a good way for evolution to handle the problem.

This isn't a timeout issue.  Evolution's response to a timeout is
different: it puts up a sodding great big error banner and retries later
(and never removes the error banner, even on later success), but at
least it doesn't prompt for a password.

These errors from Yahoo are internal server errors, so the login is
successful, but the servers can't cope with something and throw back an
error.  Unfortunately it looks like Evolution's error handling in such a
situation is fairly broad-brushed and the generic response is to ask for
a password.

So the underlying problem is from Yahoo, but Evolution's response to the
problem is not as helpful as it could be.

But, to inject a bit of controversy into this, is the usage of the very
old POP protocol sufficient that the devs should spend time looking at
this sort of issue?  Why do people still use POP, when IMAP is a more
complete solution and things like offline-imap and fetchmail exist for
those who work in an offline environment?

P.