e798b91840
getifaddrs() can return entries where ifa_addr is NULL. Check for this
...
before accessing anything in ifa_addr.
ok claudio@
2020-12-30 18:41:06 +00:00
c9d6433d50
briefly mention /etc/examples/ in the FILES section of all the
...
manual pages that document the corresponding configuration files;
OK jmc@, and general direction discussed with many
2020-02-10 13:18:20 +00:00
df69c215c7
When system calls indicate an error they return -1, not some arbitrary
...
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.
2019-06-28 13:32:41 +00:00
18d15a648c
struct sockaddr_storage is guaranteed to be large enough, no need to check
...
the size.
OK kn, deraadt
2018-10-15 11:29:27 +00:00
f414793931
Use <fcntl.h> instead of <sys/file.h> for open() and friends.
...
Delete a bunch of unnecessary #includes and sort to match style(9)
while doing the above cleanup.
ok deraadt@ krw@
2018-04-26 12:42:50 +00:00
276bf740fc
Fix pledge(2) usage
...
We should not try to tighten pledge promises in the master inetd
process, since those are needed to reload the configuration. Only
tighten down if 1) we have forked 2) we're running one of the builtin
routines, which are dealing with network input.
Pointed out by and ok benno@
2017-10-04 23:56:48 +00:00
53213b25b8
Don't try to handle IPv4-compatible IPv6 addresses
...
ok florian@ millert@
2017-10-04 20:15:52 +00:00
c9d10ebe31
inetd actually supports stream and datagram sockets only
...
ok deraadt@
2017-10-04 19:52:26 +00:00
942ef79866
inetd needs cpath pledge to unlink unix socket at program termination.
...
From Anthony Coulter - thanks!
ok jca@
2017-10-04 18:08:26 +00:00
9e23c5a985
remove lint comments
2016-08-25 05:23:19 +00:00
faa33665ab
for some time now mandoc has not required MLINKS to function
...
correctly - logically complete that now by removing MLINKS from base;
authors need only to ensure there is an entry in NAME for any function/
util being added. MLINKS will still work, and remain for perl to ease
upgrades;
ok nicm (curses) bcook (ssl)
ok schwarze, who provided a lot of feedback and assistance
ok tb natano jung
2016-03-30 06:38:40 +00:00
4c67a02ce5
Remove a (now) obsolete comment about shortened IPv4 syntax
...
OK deraadt@ jca@
2016-03-16 20:16:39 +00:00
3c0f199211
Replace strncpy(3) calls for safer strlcpy(3)
...
OK deraadt@ jca@
2016-03-16 20:15:15 +00:00
2eba59a5c7
Replace atoi(3) calls for safer strtonum(3)
...
OK deeradt@ jca@
2016-03-16 20:09:58 +00:00
626fd955e6
remove a NULL-check before free()
2015-12-22 19:44:01 +00:00
e488283d8a
these files document their config files too: adjust NAME accordingly;
...
ok schwarze
2015-11-14 01:22:04 +00:00
b1e9a47392
needs pledge "cpath" to allow unlink of AF_UNIX sockets eventually
2015-11-01 19:59:28 +00:00
4359013679
inetd needs pledge "unix" to bind AF_UNIX sockets
2015-11-01 00:31:18 +00:00
362920ef3c
pledge "abort" left behind accidentally
2015-10-22 12:34:25 +00:00
d4cb28ceff
optarg and optind are declared by unistd.h
2015-10-19 11:52:51 +00:00
2448817a8f
Add err.h, missing after pledge() introduction.
2015-10-19 11:51:17 +00:00
24ecfc683a
Make use of pledge(2); initial diff from deraadt@
...
As Theo says, there's probably room for stricter pledge requests, but
this would involve refactoring.
2015-10-18 16:11:19 +00:00
ded2548f1c
actually, it uses getaddrinfo
2015-10-18 15:28:03 +00:00
e34c05504d
reduce .Nd to one line and kill .Tn while here
2015-09-10 15:16:43 +00:00
b9fc9a728f
Replace <sys/param.h> with <limits.h> and other less dirty headers where
...
possible. Annotate <sys/param.h> lines with their current reasons. Switch
to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change
MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where
sensible to avoid pulling in the pollution. These are the files confirmed
through binary verification.
ok guenther, millert, doug (helped with the verification protocol)
2015-01-16 06:39:28 +00:00
8d1bc828ab
replace the use of select and hand rolled deferral of signal handling
...
with libevent.
this has been in snaps for a while without any negative (or positive)
feedback.
review and tweaks by bluhm@ nicm@
ok bluhm@ deraadt@
2014-10-29 03:33:14 +00:00
c686248187
jmc@ told me i broke the documentation. inetd doesnt write a pidfile so
...
we shouldnt say it does.
2014-10-13 01:56:44 +00:00
2f2695c65f
dont need to store the same debug flag in two separate variables.
...
ok millert@ deraadt@
2014-10-13 01:53:14 +00:00
d0ca4f7416
ansi function declarations.
...
ok gcc+sha256
2014-10-12 09:36:42 +00:00
b3815cd84d
cut out writing a pidfile. rc.d doesnt rely on it, and nor should you.
...
both bcook@ and schwartz@ suggested removing the pidfil unlink and path
ok sthen@ schwartz@
2014-10-12 08:55:25 +00:00
a31c5aa33f
Convert malloc/memset to calloc, and remove pointless casts while here.
...
From Benjamin Baier.
2014-06-17 03:12:37 +00:00
fd49663c75
remove references to rshd;
2014-04-18 22:04:54 +00:00
af5650ac29
obvious unsigned char casts for ctype
2013-11-23 17:24:29 +00:00
3e37b709e8
missing protos
2013-11-12 19:44:44 +00:00
382358551f
for datagrams, assume other protocols should fail
...
ok millert
2013-04-19 18:03:16 +00:00
62e3c25282
handle ECONNABORTED errors from accept(). In many code blocks they can be
...
ignored silently and without aborting, much like EINTR and EWOULDBLOCK are.
ok's from various maintainers of these directories...
2013-03-11 17:40:09 +00:00
f95242c2ac
small cleanup(%hd -> %d)
...
yes deraadt@
2012-04-19 14:33:24 +00:00
3c4281b152
clean up some newly created lies now that tftpd doesnt run out of inetd.
...
poke by deraadt@
2012-03-06 01:25:46 +00:00
b9e0f9e667
replace a few remaining UNIX terms with .Ux macros; this diff
...
completes the changes from Oct 22.
2009-12-29 10:25:43 +00:00
51a9e7d759
we have closefrom(2), so use it; ok deraadt@ tedu@
2009-11-02 20:03:01 +00:00
043fbe51c1
rcsid[] and sccsid[] and copyright[] are essentially unmaintained (and
...
unmaintainable). these days, people use source. these id's do not provide
any benefit, and do hurt the small install media
(the 33,000 line diff is essentially mechanical)
ok with the idea millert, ok dms
2009-10-27 23:59:19 +00:00
1b830f879d
Do not use __progname in inetd as it is not portable. Instead hardcode it to
...
"inetd" in the few locations that print out the program name.
requested and OK deraadt@
2008-07-28 15:42:07 +00:00
2c4a6bc86b
Use __progname instead of hand rolled version using argv[0].
...
Diff from Gleydson Soares gsoares at gmail dot com
OK millert@
2008-07-27 10:06:10 +00:00
4463de31c1
add an underscore to argument name.
2008-06-28 10:54:45 +00:00
ffce9674b3
adderss -> address
2008-06-04 18:11:34 +00:00
3ca821ab96
validate argv, okay otto@, henning@
2008-01-25 13:52:20 +00:00
e7a08e4a39
remove out of date/irrelevant section of BUGS dealing with ip6;
...
ok deraadt
2008-01-06 16:04:34 +00:00
91107b6999
- document what happens when a service exceeds its max spawn rate. this is
...
based on a diff from reed@netbsd, but i chose a different fix
- whilst here, make the descriptions of -R and .max more consistent
ok sthen sobrado(in part, at least)
2008-01-06 13:49:42 +00:00
844173ee7c
sync usage() w/ SYNOPSIS;
2008-01-05 09:53:42 +00:00
673a9bc797
use strcspn to properly overwrite '\n' in fgets returned buffer
...
ok pyr@, ray@, millert@, moritz@, chl@
2007-09-11 16:30:59 +00:00
benno
schwarze
deraadt
florian
guenther
jca
tedu
jmc
mestre
mmcc
dlg
lteo
gsoares
sobrado
otto
claudio
miod
espie
gilles