[Tilesathome] Regression: Repeated errors don't lead to client exiting

[Matthias Julius]

Dirk-Lüder Kreie <osm-list at deelkar.net> writes:

> Hi,
> there used to be a check for repeated errors of the same kind, so that
> one client with some problem that would not go away by trying another
> tile would not hog the entire queue.
>
> Looking at the logs, this does not seem to be the case anymore.
> It seems that on forked rendering the client can produce as many
> rendering errors in a row (like black tiles) as it wants and the
> client will never stop trying the next tile and the next and so on,
> ever failing.
>
> This needs to be remedied, best by someone that understands the
> current exception based error handling.

Yes, this stuff has been neglected lately and certainly needs some
overhaul.  I will take a look once I get around to it.

Matthias