CVS log for src/libexec/telnetd/telnetd.c

Up to [cvs.NetBSD.org] / src / libexec / telnetd

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

Pull up following revision(s) (requested by riastradh in ticket #1927):

	lib/libtelnet/encrypt.h: revision 1.10
	libexec/telnetd/telnetd.c: revision 1.59
	libexec/telnetd/sys_term.c: revision 1.50
	lib/libtelnet/encrypt.c: revision 1.20
	lib/libtelnet/enc_des.c: revision 1.18
	lib/libtelnet/enc-proto.h: revision 1.11
	libexec/telnetd/state.c: revision 1.33
	libexec/telnetd/state.c: revision 1.35
	libexec/telnetd/telnetd.c: revision 1.60

Fix memory leak - free resources allocated by getaddrinfo

Fix off by one in telrcv()

In case of "\r" in the data buffer, the code was unconditionally looking ahead
to next character, even if "\r" was last character in the buffer. That
condition leads to read outside of the data (one byte after the array)
Thanks christos@ for the review
PR bin/58787  libtelnet - avoid using a global variable


From RVP -- make 'encrypt_debug_mode' be a static (file scope)
variable instead of extern, and convert the (had been recently
anyway) unused function encrypt_debug() into an accensor function
for it.

This is the minor (almost irrelevant) change from PR bin/58787
PR bin/58787  telnetd - handle auto authentication better
Apparently from FreeBSD via RVP -- but FreeBSD deleted telnetd
more than 2 years ago, so I assume instead from

        https://github.com/cschuber/freebsd-telnet/tree/main/contrib/telnet

Pull up following revision(s) (requested by riastradh in ticket #1013):

	lib/libtelnet/encrypt.h: revision 1.10
	libexec/telnetd/telnetd.c: revision 1.59
	libexec/telnetd/sys_term.c: revision 1.50
	lib/libtelnet/encrypt.c: revision 1.20
	lib/libtelnet/enc_des.c: revision 1.18
	lib/libtelnet/enc-proto.h: revision 1.11
	libexec/telnetd/state.c: revision 1.33
	libexec/telnetd/state.c: revision 1.35
	libexec/telnetd/telnetd.c: revision 1.60

Fix memory leak - free resources allocated by getaddrinfo

Fix off by one in telrcv()

In case of "\r" in the data buffer, the code was unconditionally looking ahead
to next character, even if "\r" was last character in the buffer. That
condition leads to read outside of the data (one byte after the array)
Thanks christos@ for the review
PR bin/58787  libtelnet - avoid using a global variable


From RVP -- make 'encrypt_debug_mode' be a static (file scope)
variable instead of extern, and convert the (had been recently
anyway) unused function encrypt_debug() into an accensor function
for it.

This is the minor (almost irrelevant) change from PR bin/58787
PR bin/58787  telnetd - handle auto authentication better
Apparently from FreeBSD via RVP -- but FreeBSD deleted telnetd
more than 2 years ago, so I assume instead from

        https://github.com/cschuber/freebsd-telnet/tree/main/contrib/telnet

PR bin/58787  telnetd - handle auto authentication better

Apparently from FreeBSD via RVP -- but FreeBSD deleted telnetd
more than 2 years ago, so I assume instead from
	https://github.com/cschuber/freebsd-telnet/tree/main/contrib/telnet

Fix memory leak - free resources allocated by getaddrinfo

Pull up following revision(s) (requested by dholland in ticket #1757):

	libexec/telnetd/telnetd.c: revision 1.58 (via patch)

telnetd: fix the crash that's been talked about the past couple days.

(Move initialization of the slc table earlier so it doesn't get
accessed before that happens.)

Calling the crash a DoS or security problem is a bit overwrought; it's
just a bug.

Pull up following revision(s) (requested by dholland in ticket #1508):

	libexec/telnetd/telnetd.c: revision 1.58 (via patch)

telnetd: fix the crash that's been talked about the past couple days.

(Move initialization of the slc table earlier so it doesn't get
accessed before that happens.)

Calling the crash a DoS or security problem is a bit overwrought; it's
just a bug.

telnetd: fix the crash that's been talked about the past couple days.

(Move initialization of the slc table earlier so it doesn't get
accessed before that happens.)

Calling the crash a DoS or security problem is a bit overwrought; it's
just a bug.

s/desireable/desirable/ in comments.

Mostly merge changes from HEAD upto 20200411

Pull up following revision(s) (requested by kamil in ticket #61):

	libexec/telnetd/telnetd.c: revision 1.56
	libexec/telnetd/sys_term.c: revision 1.49

telnetd: Stop defining the same variables concurrently in bss and data
auth_level and require_secure_login were defined in two places:

 - global.c that picked it from headers and removed 'extern'
 - telnetd.c that initialized it to 0

line was defined twice: in global.c and sys_term.c with a non-zero value.

Remove the definition and initialization from sys_term.c and initialize
the variable early in main().

Detected during the build of telned with Address Sanitizer (MKSANITIZER).
Reviewed by <mrg>

telnetd: Stop defining the same variables concurrently in bss and data

auth_level and require_secure_login were defined in two places:

 - global.c that picked it from headers and removed 'extern'
 - telnetd.c that initialized it to 0

line was defined twice: in global.c and sys_term.c with a non-zero value.
Remove the definition and initialization from sys_term.c and initialize
the variable early in main().

Detected during the build of telned with Address Sanitizer (MKSANITIZER).

Reviewed by <mrg>

Rebase to HEAD as of a few days ago.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

Global variable is never a null pointer.

sync with head

Use __dead

PR/45805: Borodin Oleg: telnet/telnetd with -DAUTHENTICATION -DENCRYPTION
(rfc2941)
- separate MKCRYPTO from USE_KERBEROS
- WARNS=4

#include <com_err.h> -> #include <krb5/com_err.h>.

Sync with wrstuden-revivesa-base-2.

Remove the \n and tabs from the __COPYRIGHT() strings.

change (mostly) int to socklen_t.  GCC 4 doesn't like that int and
socklen_t are different signness.

Make the debug variable static, to accomodate static linking with libssh.a.

Apply patch (requested by lha in ticket #2041):
Remove static local variable so its easier to correctly use strlcpy.

Apply patch (requested by lha in ticket #2041):
Remove static local variable so its easier to correctly use strlcpy.

Pull up revision 1.48 (requested by lha in ticket #502):
Remove static local variable so its easier to correctly use strlcpy.

Remove static local variable so its easier to correctly use strlcpy.

Pull up revision 1.47 (requested by lukem in ticket #400):
Ensure IF is initialized before trying to use it.
Found with -Wuninitialized.

Ensure IF is initialized before trying to use it.
Found with -Wuninitialized.

de-__P, ANSIfy function declarations, remove obsolete "register" declarations.

PR/28473: Jed Davis: telnetd loses data and corrupts crypto state due to
typo in use of poll(). There was an && instead of an & (hi charles)

Pass -a [srrvis'd struct sockaddr] to login, so that login can record the
address where we have logged in from.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22284, verified by myself.

Since telnetd has ignored the USER environment variable for logging in for
quite a bit of time, make telnetd ignore it completely now.  This results
in the :if=: entry in the default gettytab entry to honored instead of
being ignored.  The -h option to telnetd will continue to suppress the
inclusion of :if=:

use bounded string op.  bump shlib major for libtelnet due to API change.

SIGINFO is defined

remove unneeded #ifdefs.  improves readability.
(confirmed that exact same *.o is produced)

use bounded string op (there still are remains)

use strlcpy

PR/21233: Onno van der Linden: -U option does not work as advertised.

Remove dead code.

select() -> poll()

add some more ifdefs, so that we don't define variables we don't use.

implement -u <len>. PR 13676

-edebug should only be mentioned in the help #ifdef ENCRYPTION, not #ifdef
AUTHENTICATION. Noted (with patch) by Brian Ginsbach.

Remove code #ifdef'd on NEWINIT (never used on NetBSD).
Sync with Cray, patch by Brian Ginsbach in private mail.

Remove lots of #ifdef's for dead Cray hardware and obsolete UNICOS releases;
also some Cray-specific fixes. No change for NetBSD.
From Brian Ginsbach of Cray Inc. in bin/13614 and private mail.

Fix typo, and duplicate ';'. From bin/13667 by Brian Ginsbach.

Add some (const char *) casts per bin/13667 by Brian Ginsbach.

Pull up revision 1.26 (requested by itojun):
  Security-related fixups for telnet
    Bound-check all "*nfrontp++" code.  From OpenBSD and Heimdal.
    Complain if file descriptor goes above FD_SETSIZE.  From OpenBSD.

bound-check all "*nfrontp++" code.  from openbsd/heimdal.
bark if file descriptor goes above FD_SETSIZE.  from openbsd.

XXX needs more checking.
XXX what is tab stop size for this code?  need more consistency...
XXX we should really remove #ifdef CRAY, UNICOS5 and such.
we just cannot read it through.

fix redundant declarations.

use LOG_ERR for fatal errors, LOG_WARNING for other errors
minor knf
deprecate LOG_ODELAY; it's the default

Buffer overflow and environment passing fixes [originally inspired from
FreeBSD]. We really didn't need the environment fixes, but for the sake
of keeping the code as close as possible...

make error case handling in -daemon case meet with other portion of the code.
(use fprintf-then-exit, not errx - costmetic)

Update from trunk:
Bring the telnet situation back into better shape.  Specifically,
pull in just about all of the differences from the crypto-us telnet
suite (which includes Kerberos 4 and connection encryption support).
Also bring in the Kerberos 5 support from the Heimdal telnet, and
frob a little so that it can work with the non-Heimdal telnet suite.

There is still some work left to do, specifically:
- Add Heimdal's ticket forwarding support to the Berkeley Kerberos 4
  module.
- Add connection encryption support to the Heimdal Kerberos 5
  module.  Hints on this can be taken from the MIT Kerberos 5
  module which still exists in crypto-us.

However, even with the shortcomings listed above, this is a
better situation than using the stock Heimdal telnet suite,
which does not understand the IPSec policy stuff, and is also
based on much older code which contains bugs that we have already
fixed in the NetBSD sources.

Bring the telnet situation back into better shape.  Specifically,
pull in just about all of the differences from the crypto-us telnet
suite (which includes Kerberos 4 and connection encryption support).
Also bring in the Kerberos 5 support from the Heimdal telnet, and
frob a little so that it can work with the non-Heimdal telnet suite.

There is still some work left to do, specifically:
- Add Heimdal's ticket forwarding support to the Berkeley Kerberos 4
  module.
- Add connection encryption support to the Heimdal Kerberos 5
  module.  Hints on this can be taken from the MIT Kerberos 5
  module which still exists in crypto-us.

However, even with the shortcomings listed above, this is a
better situation than using the stock Heimdal telnet suite,
which does not understand the IPSec policy stuff, and is also
based on much older code which contains bugs that we have already
fixed in the NetBSD sources.

Make 'if' capability from gettytab work. *shudder*.

Pull up to last week's -current.

do not touch __ss_len directly, they are not supposed to be touched.

Nuke some whitespace.

Put the RCS ID in the correct place.

dual-stack telnetd.

Fix for PR 5821
- s/SecurID/SECURELOGIN/g and define SECURELOGIN in Makefile
- update man page to reflect S/Key and Kerboros authentication used
  not SecurID

Declare ints, too, don't default them.

- use an array MAXHOSTNAMELEN+1 size to hold hostnames
- ensure hostname from gethostname() is nul-terminated in all cases
- minor KNF
- use MAXHOSTNAMELEN over various other values/defines
- be safe will buffers that hold hostnames

No need to declare optind and optarg.

don't include <sys/cdefs.h> twice, kill old-style getpty() declaration

use <termcap.h>

WARNS?=1.  RCS ids.

Put a whitespace after the keyword `switch'.

Compare a return value of getopt() against -1 instead of EOF.

apply Thorsten's patch to add gettytab support to telnetd.

RCS id cosmetics.

update to telnetd from 95.10.23 version.  Includes Borman's scrub_env()
to keep stuff out of the environment on the way to exec (we already have
one in the state machine that keeps them from arriving over the wire, so
this should be redundant, but it'll make any further updates easier to
have it present).

also, RCS Id police.

Import original 4.4-Lite version.

new telnetd from ftp.cray.com.  Encryption support ripped out, pending
figuring out what to do about it...

Add RCS identifiers.

initial import of 386bsd-0.1 sources

Initial revision