| Age | Commit message (Collapse) | Author |
|---|
|
- add cronjob for splash_sync (every minute for debugging; to be
increased)
- add redirection for unsplashed users in queen.enter, remove it in
queen.leave
- add /etc/splash.sh to centralize functionality used in
/sbin/splash_sync and /www/service/*splash*
- make splash_sync to use a transactional update approach: insert
iptable rules for all splashed MACs into a new chain, then replace
old and new chain; this avoids downtime even if splash_sync is slow
- remove allowed services for splashed users; this is intended to not
move old cruft over to new code; if any of the those services shall
be reenabled, it should happen in /etc/firewall.user
- make /sbin/splash_sync aware of the current state of FSM inetable
(read via /sbin/fsm with optional command line override); this
allows to unconditionally call splash_sync via cron
- add locks to all places where the splash state is touched
- remove multiple escaping/quoting/encoding problems in splash cgi
scripts
- reenable link to the page the user initially demanded in
/www/service/cgi-bin/splash_click
|
|
|
|
|
|
For debugging reasons both cronjobs are executed once per minute. This
rate should be reduced before deployment in larger networks.
|
|
The table broadcasted once per minute over br-mesh. To fill the table
with contents on all nodes, run:
p2ptbl update /tmp/p2ptbl/test somekey somevalue
on one node. If the command shall have immediate effect, run:
p2ptbl update /tmp/p2ptbl/test somekey somevalue br-mesh
|
|
|
