[Tile-serving] [osm2pgsql-dev/osm2pgsql] Allow a limit to be set on expiry of line segments (PR #2185)

[Sarah Hoffmann]

Generally speaking, I'd be okay with accepting this as an operational quick-fix. The code is minimally invasive and in a part of osm2pgsql that's on the way out anyway. So we don't have to maintain that stuff forever.

My main concern here is with joto's point 4. The change not only prevents expiry when the vandalism occurs but also when the vandalism is fixed. This may not be a much of a problem on the server with average load. However, especially on the osm.org servers with their very high load, this may lead to a lot of stale tiles where the vandalism remains visible because another change caused the tiles to be rerendered while the vandalism was active. Opinions?

I wonder if it wouldn't be better to attack the problem a bit earlier and not allow to lines with very long segments to be written to the output tables. Sadly, this is only possible with the flex output right now, so we'd need the osm-carto port for flex first. But once done it can be flexibly handled in style, so that a sensible maximum length can be set depending on the type of object (allow more for administrative boundaries, less for highways).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/osm2pgsql-dev/osm2pgsql/pull/2185#issuecomment-2136983978
You are receiving this because you are subscribed to this thread.

Message ID: <osm2pgsql-dev/osm2pgsql/pull/2185/c2136983978 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tile-serving/attachments/20240529/1965aad1/attachment.htm>