If we are a master and we don't have peers, there is no use keeping the
sessions alive, when we'll restart we will have forgotten them all.
Better gracefully close them, so they can be quickly respawn on restart,
instead of having to wait for timeouts.
Bump heartbeat version to handle Class entry in session (v6).
Re-arrange session struct to remove padding.
Update cluster code to handle v6 packets. Drop compatability for pre-v5.
- Call sessionshutdown() when a tunnel is dropped rather than
sessionkill() to ensure that RADIUS stop records are sent.
- Cleanup: make a bunch of global functions/variables static.
- fix for LASTSEEN breakage: don't do anything in the CLI other than
flag changes to be made by the parent
- split out master parts from cluster_check_master() into cluster_check_slaves()
- add assertions to help identify odd LASTSEEN breakage
- make cluster_hb_interval work; include interval/timeout in heartbeats
so that a change on the master is propagated immediately to the slaves
- use fast heartbeats when there are slaves not up to date
- ensure basetime of shut down master is set to zero (prevent delayed election)
- fix radius session leak on IPCP timeout
- fix some off-by-one errors in tunnel/session loops
- Major release
- Completely replace active/standby clustering with a new peer-to-peer
clustering method which allows much greater throughput and is a lot more fault
tolerant
- Add internal tbf implementation for throttling without relying on tc and
kernel HTB
- Add support for iBGP and eBGP to advertise routes
- Add cli commands "show cluster", "show bgp", "show ipcache", "show throttle",
"show tbf", "suspend bgp", "restart bgp", "show user"
- Interception destination must be set per-user
- If SMP machine, allow use of SCHED_FIFO, which should improve performance
- Added config option to send GARP at startup
- Added plugin_become_master and plugin_new_session_master plugin hooks
- Remove useless sessionsendarp(). This isn't needed now that we are using TUN
instead of TAP.
- ICMP rate limiting so not every unreachable packet is replied with an ICMP
unreachable message
- mangle table is not required on anything but the cluster master, so slaves
will drop the mangle table and attempt to unload the ip_conntrack module
- Statically assigned IP addresses (by Radius) work now
- Add -d command-line flag to detach and become a daemon
- Configuration file is now "/etc/l2tpns/startup-config"
- Reduced MIN_IP_SIZE to 0x19 to stop a pile of Short IP warnings
- Resend initial IPCP request until it's acknowleged by the client
- Better radius session cleanup logic
- Many miscellaenous bugfixes and performance enhancements
- Thanks to Michael O'Reilly and Brendan O'Dea for most of these new features
- Change all strcpy() calls to strncpy() to avoid buffer overflow potential
- Add ICMP host unreachable support
- Logging to syslog if log_file = "syslog:facility"
- Now requires libcli 1.5
- All configuration moves to a config structure
- Ability to modify and write config on the fly through command-line interface
- Config file support is removed, and now handled by the cli
- Show hostname in cli prompt
- Keep current state type for tunnels
- Add uptime command do CLI, which also shows real-time bandwidth utilisation
- Add goodbye command to cluster master, which forces droppping a slave
- Cache IP address allocation, so that reconnecting users get the same address
- Fix tunnel resend timeouts, so that dead tunnels will be cleaned up
- Allocate tunnels and radius without using a linked list which had issues
- Fix some off-by-one errors in tunnel and session and radius arrays
- Save and reload ip address pool when dieing
- Check version and size of reloaded data when restarting
- Remove plugin_config support
- Remove old support for TBF which didn't work anyway. HTB is required to do throttling now.
- Add COPYING and Changes files
