summaryrefslogtreecommitdiff
path: root/sbin
AgeCommit message (Collapse)Author
2002-12-06let pfctl -v print out "block drop" instead of just "block" in the dropHenning Brauer
case. fixes the last self-regress test. ok markus@
2002-12-06Section and tag comparisions should be case-insensitive. PR#3010, Mike Neuman.Hakan Olsson
2002-12-06unkown; torh@bogus.netTheo de Raadt
2002-12-06Introduce anchors and named rule sets, allowing to load additional ruleDaniel Hartmeier
sets with pfctl and evaluate them from the main rule set using a new type of rule (which will support conditional evaluation soon). Makes maintenance of sub-rulesets simpler for pfctl and daemons. Idea and ok deraadt@
2002-12-05fix a small cosmetic problem. if ifa_lookup was called in LOOKUP_NET mode,Henning Brauer
it would return things like 127.0.0.1/8. when 127.0.0.1/8 is passed through the new selftests, it becomes 127.0.0.0/8. fix: instead of just copying addr.mask in ifa_lookup, use set_ipmask, so it is 127.0.0.0/8 from the beginning on.
2002-12-05one step further to "program output is valid program input" and the relatedHenning Brauer
extra regress testing opportunities (an excellent idea by markus@). only print rule numbers in case of pfctl -v -v; but not with pfctl -v. ok markus@ frantzen@ mcbride@
2002-12-05-quote varset in -v outputHenning Brauer
-don't add 's' in set timeout -v output makes the feedback regression tests succeed. ok markus@ mcbride@
2002-12-05Replace warnx with warn on ioctls so that we get the error message associatedRyan Thomas McBride
with the warning. From DJ Gregor. ok henning@ dhartmei@
2002-12-05correct sizeof() to memset(). sync w/kameJun-ichiro itojun Hagino
2002-12-04bit_ffc(3) returns a -1 location for a full bitset. Problem found byHakan Olsson
Olivier Courtay.
2002-12-04allow ACQUIRE with empty EXT_ADDRESS_SRC; PR 2996; ok angelos@, ho@Markus Friedl
2002-12-04in host(), apply a given netmask to all members of the node_host listHenning Brauer
returned by ifa_lookup(). reported by Alejandro G. Belluscio ok dhartmei@
2002-12-04do not warn for unused variables except with -v -v; henning okTheo de Raadt
2002-12-03Add -4/-6 cmdline options to select what address family (IPv4,v6) to use.Hakan Olsson
niklas@ ok.
2002-12-03LOG_DBG for missing ca/, certs/, crls/ dirs, notHakan Olsson
log_error(). Suggested by markus@.
2002-12-03be nice and shutdown the socket after beeing done; pointed out by kenjiroHenning Brauer
2002-12-02KNFHenning Brauer
2002-12-02permit more flexible pass/block lines. the options giving at the end ofTheo de Raadt
a line can now be placed in any order, and some merging of duplicat options is also done. this is easier to use.
2002-12-02allow for "altq on $interface" without specifying a bandwidth. TheHenning Brauer
interface's bandwidth is taken then. as a side-effect, "altq on $interface bandwidth xx%" is also possible now, it's relative to the interface's bandwidth then. ok theo
2002-12-01deal with not-ALTQ-enabled kernelsHenning Brauer
2002-12-01Match pf_ioctl.c cleanup; set up the pool_ticket when loading rulesRyan Thomas McBride
ok dhartmei@ henning@
2002-11-30oopsTheo de Raadt
2002-11-30do not syslog with extra newlineTheo de Raadt
2002-11-30move damn unmask() proto into place as wellMichael Shalayeff
2002-11-30move unmask back into pf_print_state.c where it was, and please keep it ↵Michael Shalayeff
there; henning@ ok
2002-11-29no need for extra hrs; henning@ okMichael Shalayeff
2002-11-29coredumps are not nice.Henning Brauer
deal with the fact that when we are merging the lists in host_list both can be NULL. found at EuroBSDCon 2002 while I was explaining the expansion process to Paul de Weerd ok dhartmei@
2002-11-29in pfctl_find_altq_node also compare the interface.Henning Brauer
fixes pfctl -sq output in case the queuespec was expanded to >1 queues on different interfaces
2002-11-29print queuespec only once in verbose mode, even if referenced more oftenHenning Brauer
2002-11-29FREE_LIST nqueues later; can be used more than onceHenning Brauer
2002-11-28qlimit before tbrsizeHenning Brauer
2002-11-28print qlimit on altq if not equal to the default valueHenning Brauer
2002-11-28allow qlimit on altq specHenning Brauer
2002-11-28- Remove extra space before -> in print_rdrRyan Thomas McBride
- Print hex keys lowercase ok dhartmei@ henning@
2002-11-28- Adjust to match replacement of md5 with pf_hashRyan Thomas McBride
- Allow keys to be specified in hex (must be 128 bits, mainly to make pfctl -v output paresable by pfctl) or as a string. Randomly generate key if one is not specified. Suggestions and fixes from camield@ ok mickey@ camield@ henning@
2002-11-27handle the "no bandwidth specified" case earlier and easier.Henning Brauer
2002-11-27-use a #define for default qlimit instead of hardcoding itHenning Brauer
-print qlimit in print_queue if it is not the default one -assign the default qlimit early in the parser instead of using 0
2002-11-27next round in print fixes:Henning Brauer
-only print priority if it is not equal to the default one -space handling overhaul
2002-11-27use a #define for default priority instead of hardcoding itHenning Brauer
2002-11-27Make sure daemon() does not close any fd opened during initialization.Hakan Olsson
niklas@ ok.
2002-11-27in print_altq() correctly print scheduler type and optionsHenning Brauer
2002-11-27fix tbrsize specHenning Brauer
2002-11-27Update document date.Hakan Olsson
2002-11-27fix error message for altq ifspec without child queuesHenning Brauer
2002-11-26make bandwidth specification on queue truly optionalHenning Brauer
2002-11-26queue comes before NATHenning Brauer
2002-11-25repair decide_address_familyHenning Brauer
you cannot just taked the first address family you meet as rule's address family... either all are equal, or the rule has no explicit AF. found by danh@ ok theo
2002-11-25for loN and link1 translate the iface into address properly; henning@ okMichael Shalayeff
2002-11-251Kb = 1000b, not 1024; as discussed with kjc@ and theoHenning Brauer
2002-11-25KNFHenning Brauer