summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorhttp://schmonz.livejournal.com/ <http://schmonz.livejournal.com/@web>2008-07-30 01:25:05 -0400
committerJoey Hess <joey@kitenet.net>2008-07-30 01:25:05 -0400
commitdd25c7c4afa8f57e909fed63fb6bcf1648de531b (patch)
tree4c3d59e2f30bd740924c38bda5b42660f2e70354
parentae04ca55467ff3c672a423caedf0daff5a5a8a47 (diff)
-rw-r--r--doc/plugins/contrib/unixauth/discussion.mdwn3
1 files changed, 3 insertions, 0 deletions
diff --git a/doc/plugins/contrib/unixauth/discussion.mdwn b/doc/plugins/contrib/unixauth/discussion.mdwn
index 91c59ff1d..863e3c91a 100644
--- a/doc/plugins/contrib/unixauth/discussion.mdwn
+++ b/doc/plugins/contrib/unixauth/discussion.mdwn
@@ -20,3 +20,6 @@ Thanks for the comments. That's definitely an undesirable interaction between pw
-- [[schmonz]]
> Have you considered using [[plugins/httpauth]] and then the appropriate apache module? There are apache modules like [mod_authnz_external](http://unixpapa.com/mod_auth_external.html) that might help. The advantage of these solutions is that they usually make the security implications explicit. -- Will
+
+Actually, yes. That's how I made sure I had pwauth working to begin with. I'm partial to the form-based approach because I'm not aware of any way to reliably "log out" browsers from HTTP authentication. If that *is* reliably possible, then I worked way too hard for no reason. ;-)
+-- [[schmonz]]