| 1 | * Planned for the future |
| 2 | |
| 3 | Netlink device that implements an Ethernet bridge. |
| 4 | |
| 5 | Modular transform code: choice of block ciphers, modes, sequence |
| 6 | numbers / timestamps, etc. similar to IWJ's udptunnel |
| 7 | |
| 8 | Path-MTU discovery for each tunnel, and fragmentation/DF support in |
| 9 | netlink code. |
| 10 | |
| 11 | * New in version 0.1.12 |
| 12 | |
| 13 | IMPORTANT: fix calculation of 'now' in secnet.c; necessary for correct |
| 14 | operation. |
| 15 | |
| 16 | (Only interesting for people building and modifying secnet by hand: |
| 17 | the Makefile now works out most dependencies automatically.) |
| 18 | |
| 19 | The netlink code no longer produces an internal routing table sorted |
| 20 | by netmask length. Instead, netlink instances have a 'priority'; the |
| 21 | table of routes is sorted by priority. Devices like laptops that have |
| 22 | tunnels that must sometimes 'mask' parts of other tunnels should be |
| 23 | given higher priorities. If a priority is not specified it is assumed |
| 24 | to be zero. |
| 25 | |
| 26 | Example usage: |
| 27 | site laptop { ... |
| 28 | link netlink { |
| 29 | route "192.168.73.74/31"; |
| 30 | priority 10; |
| 31 | }; |
| 32 | }; |
| 33 | |
| 34 | * New in version 0.1.11 |
| 35 | |
| 36 | Lists of IP addresses in the configuration file can now include |
| 37 | exclusions as well as inclusions. For example, you can specify all |
| 38 | the hosts on a subnet except one as follows: |
| 39 | |
| 40 | networks "192.168.73.0/24","!192.168.73.70"; |
| 41 | |
| 42 | (If you were only allowed inclusions, you'd have to specify that like |
| 43 | this: |
| 44 | networks "192.168.73.71/32","192.168.73.68/31","192.168.73.64/30", |
| 45 | "192.168.73.72/29","192.168.73.80/28","192.168.73.96/27", |
| 46 | "192.168.73.0/26","192.168.73.128/25"; |
| 47 | ) |
| 48 | |
| 49 | secnet now ensures that it invokes userv-ipif with a non-overlapping |
| 50 | list of subnets. |
| 51 | |
| 52 | There is a new command-line option, --sites-key or -s, that enables |
| 53 | the configuration file key that's checked to determine the list of |
| 54 | active sites (default "sites") to be changed. This enables a single |
| 55 | configuration file to contain multiple cofigurations conveniently. |
| 56 | |
| 57 | NAKs are now sent when packets arrive that are not understood. The |
| 58 | tunnel code initiates a key setup if it sees a NAK. Future |
| 59 | developments should include configuration options that control this. |
| 60 | |
| 61 | The tunnel code notifies its peer when secnet is terminating, so the |
| 62 | peer can close the session. |
| 63 | |
| 64 | The netlink "exclude-remote-networks" option has now been replaced by |
| 65 | a "remote-networks" option; instead of specifying networks that no |
| 66 | site may access, you specify the set of networks that remote sites are |
| 67 | allowed to access. A sensible example: "192.168.0.0/16", |
| 68 | "172.16.0.0/12", "10.0.0.0/8", "!your-local-network" |
| 69 | |
| 70 | * New in version 0.1.10 |
| 71 | |
| 72 | WARNING: THIS VERSION MAKES A CHANGE TO THE CONFIGURATION FILE FORMAT |
| 73 | THAT IS NOT BACKWARD COMPATIBLE. However, in most configurations the |
| 74 | change only affects the sites.conf file, which is generated by the |
| 75 | make-secnet-sites script; after you regenerate your sites.conf using |
| 76 | version 0.1.10, everything should continue to work. |
| 77 | |
| 78 | Netlink devices now interact slightly differently with the 'site' |
| 79 | code. When you invoke a netlink closure like 'tun' or 'userv-ipif', |
| 80 | you get another closure back. You then invoke this closure (usually |
| 81 | in the site definitions) to specify things like routes and options. |
| 82 | The result of this invocation should be used as the 'link' option in |
| 83 | site configurations. |
| 84 | |
| 85 | All this really means is that instead of site configurations looking |
| 86 | like this: |
| 87 | |
| 88 | foo { |
| 89 | name "foo"; |
| 90 | networks "a", "b", "c"; |
| 91 | etc. |
| 92 | }; |
| 93 | |
| 94 | ...they look like this: |
| 95 | |
| 96 | foo { |
| 97 | name "foo"; |
| 98 | link netlink { routes "a", "b", "c"; }; |
| 99 | etc. |
| 100 | }; |
| 101 | |
| 102 | This change was made to enable the 'site' code to be completely free |
| 103 | of any knowledge of the contents of the packets it transmits. It |
| 104 | should now be possible in the future to tunnel other protocols like |
| 105 | IPv6, IPX, raw Ethernet frames, etc. without changing the 'site' code |
| 106 | at all. |
| 107 | |
| 108 | Point-to-point netlink devices work slightly differently; when you |
| 109 | apply the 'tun', 'userv-ipif', etc. closure and specify the |
| 110 | ptp-address option, you must also specify the 'routes' option. The |
| 111 | result of this invocation should be passed directly to the 'link' |
| 112 | option of the site configuration. You can do things like this: |
| 113 | |
| 114 | sites site { |
| 115 | name "foo"; |
| 116 | link tun { |
| 117 | networks "192.168.73.76/32"; |
| 118 | local-address "192.168.73.76"; # IP address of interface |
| 119 | ptp-address "192.168.73.75"; # IP address of other end of link |
| 120 | routes "192.168.73.74/32"; |
| 121 | mtu 1400; |
| 122 | buffer sysbuffer(); |
| 123 | }; |
| 124 | etc. |
| 125 | }; |
| 126 | |
| 127 | The route dump obtained by sending SIGUSR1 to secnet now includes |
| 128 | packet counts. |
| 129 | |
| 130 | Point-to-point mode has now been tested. |
| 131 | |
| 132 | tun-old has now been tested, and the annoying 'untested' message has |
| 133 | been removed. Thanks to SGT and JDA. |
| 134 | |
| 135 | secnet now closes its stdin, stdout and stderr just after |
| 136 | backgrounding. |
| 137 | |
| 138 | Bugfix: specifying network "0.0.0.0/0" (or "default") now works |
| 139 | correctly. |
| 140 | |
| 141 | * New in version 0.1.9 |
| 142 | |
| 143 | The netlink code may now generate ICMP responses to ICMP messages that |
| 144 | are not errors, eg. ICMP echo-request. This makes Windows NT |
| 145 | traceroute output look a little less strange. |
| 146 | |
| 147 | configure.in and config.h.bot now define uint32_t etc. even on systems |
| 148 | without stdint.h and inttypes.h (needed for Solaris 2.5.1) |
| 149 | |
| 150 | GNU getopt is included for systems that lack it. |
| 151 | |
| 152 | We check for LOG_AUTHPRIV before trying to use it in log.c (Solaris |
| 153 | 2.5.1 doesn't have it.) |
| 154 | |
| 155 | Portable snprintf.c from http://www.ijs.si/software/snprintf/ is |
| 156 | included for systems that lack snprintf/vsnprintf. |
| 157 | |
| 158 | make-secnet-sites.py renamed to make-secnet-sites and now installed in |
| 159 | $prefix/sbin/make-secnet-sites; ipaddr.py library installed in |
| 160 | $prefix/share/secnet/ipaddr.py. make-secnet-sites searches |
| 161 | /usr/local/share/secnet and /usr/share/secnet for ipaddr.py |
| 162 | |
| 163 | * New in version 0.1.8 |
| 164 | |
| 165 | Netlink devices now support a 'point-to-point' mode. In this mode the |
| 166 | netlink device does not require an IP address; instead, the IP address |
| 167 | of the other end of the tunnel is specified using the 'ptp-address' |
| 168 | option. Precisely one site must be configured to use the netlink |
| 169 | device. (I haven't had a chance to test this because 0.1.8 turned into |
| 170 | a 'quick' release to enable secnet to cope with the network problems |
| 171 | affecting connections going via LINX on 2001-10-16.) |
| 172 | |
| 173 | The tunnel code in site.c now initiates a key setup if the |
| 174 | reverse-transform function fails (wrong key, bad MAC, too much skew, |
| 175 | etc.) - this should make secnet more reliable on dodgy links, which |
| 176 | are much more common than links with active attackers... (an attacker |
| 177 | can now force a new key setup by replaying an old packet, but apart |
| 178 | from minor denial of service on slow links or machines this won't |
| 179 | achieve them much). This should eventually be made configurable. |
| 180 | |
| 181 | The sequence number skew detection code in transform.c now only |
| 182 | complains about 'reverse skew' - replays of packets that are too |
| 183 | old. 'Forward skew' (gaps in the sequence numbers of received packets) |
| 184 | is now tolerated silently, to cope with large amounts of packet loss. |