5/30/2015 1:09:48 PM
session reflect the stale last access time
value, so the comparison accessedRecently uses the old value,
not the one from the (cached) last request.
The implementation is not pretty but with the current code,
I haven't found a better way :/
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/30/2015 7:14:37 AM
incorrect to assume that only the last access time
attribute is updated. Try to be a bit smarter instead,
and avoid updating the state for each request. The main drawback
of this approach is that the session may timeout too early
(5% of the timeout length in the worst case, i.e. 90s by default).
See d9b1b00e6fcfa3aac6c2290187d19f29a3e6bb9c and
https://github.com/killbill/killbill/issues/326.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/29/2015 2:22:49 PM
Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/29/2015 5:07:49 AM
Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/28/2015 6:57:00 AM
Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/27/2015 5:54:12 AM
fix https://github.com/killbill/killbill/issues/326.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/23/2015 11:24:15 AM
Pierre-Alexandre Meyer <pierre@mouraf.org>
|
5/19/2015 5:48:12 AM
through the code
... this is currently a naive experiment of avoiding shared SecureRandom contention
|
6/3/2015 9:14:41 PM
Plugin API
|
|
|
