summaryrefslogtreecommitdiff
path: root/sbin/isakmpd/monitor.c
AgeCommit message (Expand)Author
2010-05-10Various comment typos. 'wether' -> 'whether' (most popular), 'possiblity' ->Kenneth R Westerback
2007-08-11Do not complain about being not able to read non-existing files. MinorHans-Joerg Hoexer
2007-08-07m_priv_req_readdir(): check file type after fstat, since d_typeMarkus Friedl
2007-04-08o Kill another strerror() from a call to log_error(),Moritz Jodeit
2007-04-02Don't append the errno string in a log_error() call,Moritz Jodeit
2007-04-02When setting all signals to their default handlers, startMoritz Jodeit
2006-07-24Style; return is not a function. hshoexer@ ok.Hakan Olsson
2006-06-18clean up some gotos. Originally from Andrey Matveev <evol at onlineHans-Joerg Hoexer
2006-03-20make sure the command fifo is ready before isakmpd returns. ThisHans-Joerg Hoexer
2006-01-02clean up error paths using "goto", part 1. From Andrey Matveev <evol atHans-Joerg Hoexer
2005-12-20let isakmpd(8) remove it's pid/fifo file on exit.Moritz Jodeit
2005-11-26get rid of some leftovers from a cleanup during c2k5. From Andrey MatveevHans-Joerg Hoexer
2005-06-13Allow isakmpd to write a pid file when /var is a subdir (e.g. /usr/var)Todd C. Miller
2005-05-28Remove current state code, it's bogus. We'll redo this.Hans-Joerg Hoexer
2005-05-28do sanity checking on directoy entries. ok hshoexer@Moritz Jodeit
2005-05-28introduce new readdir implementation for the monitor.Moritz Jodeit
2005-05-28make path checking in the monitor a lot easier. ok hshoexer@Moritz Jodeit
2005-05-27Make monitor.c use unsigned lengths in messages. Makes this compileChad Loder
2005-05-27Additional paranoia. OK hshoexerChad Loder
2005-05-27o only pass signals from monitor to slave when pid is validMoritz Jodeit
2005-05-27move m_state.s directly into must_{read,write} instead of passingMoritz Jodeit
2005-05-27guarantee nul-termination in the monitor, we must. ok cloder@ hshoexer@Moritz Jodeit
2005-05-26simplify read/write between child and monitorHans-Joerg Hoexer
2005-05-24kill unneeded exit(). log_fatal() already does it. ok hshoexer@Moritz Jodeit
2005-05-24typoMoritz Jodeit
2005-05-23endpwent() not needed here, pointed out by theo.Hakan Olsson
2005-05-23endpwent() after last use. hshoexer@ okHakan Olsson
2005-05-10no need to use select()Hans-Joerg Hoexer
2005-05-05remove a LOG_DBG, suggested by moritzHans-Joerg Hoexer
2005-05-05more cleanup: use log_fatalx(), nicer handling of the socketpair, useHans-Joerg Hoexer
2005-05-03Remove SIGUSR2-related stuff as suggested by ho@.Moritz Jodeit
2005-05-03fix signal race, when child dies before the monitorMoritz Jodeit
2005-04-19missing endpwent(), change more carefully to _isakmpd privsep userHans-Joerg Hoexer
2005-04-08get rid of sysdep_sa_lenHans-Joerg Hoexer
2005-04-08keynote and policy always compiled inTheo de Raadt
2005-04-04spacing; ok cloderTheo de Raadt
2005-03-11simplify monitor main loopHans-Joerg Hoexer
2005-03-03fix typo in log msg.Moritz Jodeit
2005-02-27where possible, use bzero instead of memsetHans-Joerg Hoexer
2004-11-08rename char array realpath to pathreal to avoid naming conflict withHans-Joerg Hoexer
2004-11-08more monitor cleanup:Hans-Joerg Hoexer
2004-11-08no const for sysdep_sa_len (not yet), slipped in during previous commit...Hans-Joerg Hoexer
2004-11-08monitor cleanup: honor const, added missing inlcude, missing typecast (makesHans-Joerg Hoexer
2004-08-12Fix compiler warning on alpha.Hans-Joerg Hoexer
2004-08-08spacingTheo de Raadt
2004-06-26Narrow down privsep interface. Move pf_key_v2_open() to monitor.Hans-Joerg Hoexer
2004-06-25Narrow down privsep interface. Remove ui_init to monitor. So we can get rid ofHans-Joerg Hoexer
2004-06-24Remove some unused code.Hans-Joerg Hoexer
2004-06-21Strip away umask bits in monitor_fopen(). hshoexer@ ok.Hakan Olsson
2004-06-20To make debugging the unprivileged child process easier, make 'isakmpd -dd'Hakan Olsson