There's only a global configuration for now, blacklisting IPs that
hit each site with more than 50 requests per second or the same
page with more than 3 requests per second.
For SVN this may be too restrictive as I've seen many SVN requests
sent for example from the adhoc machine on every hour (does anyone
know if what is triggering that flood of requests? Is an SVN update
enough to do that ? It's mostly PROPFIND requests)
Another flood arrives from postgis.net, for example, and it's
likely due to the same problem:
So, let's see how things go, and we could then tweak the config
within VirtualHost sections to adapt to the service.
Logs are sent to daemon channel, so /var/log/daemon.log
On Thu, May 12, 2016 at 01:16:08PM +0200, Sandro Santilli wrote:
For SVN this may be too restrictive as I've seen many SVN requests
sent for example from the adhoc machine on every hour (does anyone
know if what is triggering that flood of requests? Is an SVN update
enough to do that ? It's mostly PROPFIND requests)
For the record: I've tried an svn checkout for the whole postgis
repository (all branches) but it only triggered 11 logged requests: