tomhughes left a comment (openstreetmap/openstreetmap-website#6476)
The historical context is that we need to ensure that people can't kill the
site by finding some way to trigger a very slow database query.
This is an ongoing problem right now with trace point queries which is what I
was hinting at above because at least sometimes the timeouts are not working a
the moment which is causing operational problems - the whole thing has long
been tricky to get right because rails has a habit of swallowing the exceptions
or transforming them into different sorts of exceptions hence things like this
and some of the logic in the timeout methods in the controllers.
Unfortunately it's very hard to debug and I suspect it depends on exactly when
the timeout fires as I haven't so far been able to reproduce the problem
locally I just know that we sometimes have database queries that have been
running for 30 minutes despite the timeout.
A database timeout doesn't quite do the same thing of course as this timeout in
principle also protects against the rails code itself taking a long time for
some reason.
--
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6476#issuecomment-3577201811
You are receiving this because you are subscribed to this thread.
Message ID:
<openstreetmap/openstreetmap-website/pull/6476/[email protected]>
_______________________________________________
rails-dev mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/rails-dev
