Age | Commit message (Collapse) | Author | |
---|---|---|---|
2019-06-28 | When system calls indicate an error they return -1, not some arbitrary | Theo de Raadt | |
value < 0. errno is only updated in this case. Change all (most?) callers of syscalls to follow this better, and let's see if this strictness helps us in the future. | |||
2019-01-22 | PF_ROUTE -> AF_ROUTE in the scattered sock()/setsockopt() calls | Kenneth R Westerback | |
where the "wrong" #define was used. ok dlg@ | |||
2018-10-31 | ifstated(8) needs to load configfile from within the main loop, but also to | Ricardo Mestre | |
reload it on SIGHUP so unveil(2) it with read permissions. Additionally all commands are exec'ed through /bin/sh instead of directly so we can just unveil(2) /bin/sh with x perms. OK deraadt@ | |||
2017-08-30 | Add support for interface departure and arrival. | rob | |
Ok jca@, benno@ | |||
2017-08-20 | Stop tracking interface indexes. Suggested by deraadt. | rob | |
Tweaks and ok jca@ "just get it in" deraadt@ | |||
2017-08-14 | Remove some dead logging code that was never used. | rob | |
Ok benno@ | |||
2017-08-08 | Consistent use of log.c, and removal of err.h include. Makes ifstated | rob | |
configtest output the same as other networking daemons. Ok jca@ | |||
2017-08-06 | Improve error checking during processing of routing messages. Handling of | rob | |
RTM_DESYNC encouraged by deraadt. ok jca@ benno@ | |||
2017-07-24 | Use pledge(2) | Jeremie Courreges-Anglas | |
Simple promises proposed by Rob Pierce after exploring more complicated privsep schemes. ok benno@ | |||
2017-07-23 | Don't bother initializing global variables to 0, since they will be in BSS. | Theo de Raadt | |
ok jca | |||
2017-07-23 | use ssize_t; ok jca | Theo de Raadt | |
2017-07-22 | No need for sys/ioctl.h any more | Jeremie Courreges-Anglas | |
2017-07-21 | Add a newline | Jeremie Courreges-Anglas | |
2017-07-21 | No need for ioctl(SIOCGIFDATA), getifaddrs(3) gives us the link state | Jeremie Courreges-Anglas | |
Noticed while reviewing a pledge(2) diff from Rob Pierce. ok florian@ | |||
2017-07-04 | rename fetch_state() to fetch_ifstate(). The word "state" is a bit | Sebastian Benoit | |
overloaded in this daemon. From Rob Pierce | |||
2017-07-03 | Remove obvious clear_config() comments and misleading state_change() comments. | Sebastian Benoit | |
Also relocate do_action() calls for the init block from change_state() to occur with the corresponding do_action() calls for the body block. From Rob Pierce ok stsp@ and i | |||
2017-07-02 | Second diff to rename additional "always" variable in struct ifsd_config. | Sebastian Benoit | |
from Rob Pierce. Requested by stsp@ | |||
2017-07-02 | Remove variable assignment in declaration and add whitespace to improve | Sebastian Benoit | |
readability. No functional change. from Rob Pierce. ok stsp@ and me | |||
2017-07-02 | Rename one of the "always" variables to "body" for improved readability. | Sebastian Benoit | |
No functional change. from Rob Pierce. ok stsp@ and me | |||
2017-06-28 | whitespace fixes, from Rob Pierce and me | Sebastian Benoit | |
2017-06-28 | usage() gets __dead attribute, from Rob Pierce | Sebastian Benoit | |
2017-06-27 | Hoist some privileged code in preparation for future work. | Sebastian Benoit | |
Based on an approach in vmd with mc146818/ns8250. diff by Rob Pierce <rob -AT- 2keys -DOT- ca> ok deraadt@ | |||
2017-06-18 | update log.c in ifstated to the same used in bgpd etc. | Sebastian Benoit | |
Done by Rob Pierce <rob AT 2keys DOT ca>, thanks | |||
2013-05-30 | setsockopt to see messages for interfaces in all routing domains again | Henning Brauer | |
instead of just the primary one. problem reported along with the right idea on how to fix it by Sean Gallagher (sean at teletech.com.au), actual fix by yours truly. Thanks Sean! ok ryan claudio | |||
2011-07-04 | LINK_STATE_IS_UP() cleanup userland part. There is no need to special | Claudio Jeker | |
case carp(4) interfaces anymore. LINK_STATE_IS_UP() almost always does the right job. OK deraadt@ henning@ This needs a -current kernel or link state may be not reported correctly. | |||
2011-03-15 | use appropriate timerclear macro - from Christiano F. Haesbaert | Okan Demirmen | |
ok mcbride@ | |||
2010-07-07 | Make ifstated cause a little less noise in /var/log/daemon. | Stefan Sperling | |
State transitions are still logged, but commands run are now printed only in debug mode (ifstated -d). ok deraadt | |||
2010-06-07 | setproctitle(NULL) is not needed, Michael Lechtermann; ok mpf | Theo de Raadt | |
2010-02-04 | remove assumption that internal IFSD_LINK* defines are the same as | Kevin Steves | |
LINK_STATE* defines in <net/if.h>, which were changed Sep 14. this fixes link state down being treated as up. ok claudio@ henning@ | |||
2009-06-25 | fix the function name in log_warn() text, it was borrowed from some | Stuart Henderson | |
previous test code of mine. | |||
2009-06-25 | Teach ifstated to use ROUTE_FILTER. Requested by claudio@. | Stuart Henderson | |
ok claudio@ henning@ | |||
2008-05-12 | Error out with usage line if additional arguments are given after the | Pierre-Yves Ritschard | |
option parsing. Found out the hard way by jdixon on ifstated. ok sobrado@, jdixon@, millert@ | |||
2007-11-27 | remove some unneeded #includes; ok mcbride@ | Kevin Steves | |
2007-10-25 | sync logging with most other daemons. | Pierre-Yves Ritschard | |
ok mcbride@, mpf@ | |||
2006-11-28 | handle the new link states as UP | Reyk Floeter | |
2006-03-16 | Kill useless arguments to libevent *_set() functions. | Ryan Thomas McBride | |
ok mpf@ | |||
2006-02-08 | - configuration file can have relative path | Camiel Dobbelaar | |
- do not suppress logging if there is no configuration yet, otherwise we cannot log that configuration parsing failed ok mpf@ mcbride@ | |||
2006-02-01 | Simplify expression code. Patch from markus@. | Marco Pfatschbacher | |
ok mcbride@ | |||
2006-02-01 | Run external tests synchronous at state changes and | Marco Pfatschbacher | |
call adjust_external_expressions() which updates the expressions for all states. Address similar state caching problems for external tests as in previous commit. Don't ignore set-state actions if they are happening right after a state change. (Based on a patch from Holger Mikolon). ok markus@, testing sturm@, ok mcbride@. | |||
2006-02-01 | If we reenter a state, it still has the old link state values cached. | Marco Pfatschbacher | |
Therefore, if it enters a state without calling scan_ifstate() (e.g. through an external-test) ifstated will do the wrong thing (tm). Change scan_ifstate() to first walk over all states and update the expressions _before_ they are evaluated. Help and ok markus@, testing sturm@, ok mcbride@. | |||
2006-01-25 | Remove gratuitous fetch_state() calls in state_change(). | Marco Pfatschbacher | |
It could race with a second incoming routing message, which would not trigger any action, because of the already updated status cache. Found by markus@. OK markus@ mcbride@ | |||
2006-01-23 | Open the routing socket before load_config() calls fetch_state(). | Marco Pfatschbacher | |
Otherwise we could loose routing messages that arrive in between. OK markus@, mcbride@ | |||
2005-07-28 | Fix two situations where ifstated was leaking zombies. | Marco Pfatschbacher | |
Tested by sturm@ OK henning@ | |||
2005-02-07 | Fix protos, and a KNF nit. From Andrey Matveev <andrushock@korovino.net> | Ryan Thomas McBride | |
2004-10-05 | Use priority instead of facility in vsyslog() | Marco Pfatschbacher | |
Pointed out by Peter Philipp. ok mcbride@ | |||
2004-09-21 | Start external tests and init commands also in the initial state. | Marco Pfatschbacher | |
Log the pid when killing an external program. Avoid a possible NULL dereference. ok mcbride@ | |||
2004-06-15 | knf | Theo de Raadt | |
2004-04-27 | various small fixes; | Jason McIntyre | |
ok mcbride@ | |||
2004-03-15 | call daemon(3) early; from tholo@; ok mcbride@ | Markus Friedl | |
2004-03-10 | Plug memory leak on SIGHUP. From Patrick Latifi. | Ryan Thomas McBride | |