| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
gcc -DHAVE_CONFIG_H -I. -I.. -DAX25_SYSCONFDIR=\""/usr/local/etc/ax25/"\" -DAX25
_LOCALSTATEDIR=\""/usr/local/var/ax25/"\" -g -O2 -Wall -MT axspawn.o -MD -MP
-MF .deps/axspawn.Tpo -c -o axspawn.o axspawn.c
[...]
axspawn.c:1049:7: warning: unused variable ‘command’ [-Wunused-variable]
char command[1024];
^
[...]
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| |\
| |
| |
| |
| | |
Conflicts:
ax25/axspawn.c
|
| | |
| |
| |
| | |
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
| |
| |
| |
| |
| |
| | |
Indentation by tabs only. Move case labels in switches are on the same
level as the switch keyword.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
| |
| |
| | |
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
| |
| |
| | |
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
| |
| |
| | |
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
| |
| |
| |
| |
| |
| | |
They only inflate the .data section of the binary. Initializations to
FALSE are still left to do.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
| |
| |
| | |
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
cpu power, if it was called from axspawn. This could be prevented if
SIGCHLD is masked.
To increase security, no longer call useradd via system() but fork() and
execve().
Fixed a bug when useradd failed then the login process continued anyway.
Empty lines in axspawn.conf are now valid.
The user_shell config variable is optional. If not specified, /bin/bash is
used. Except: if create_with_useradd is configured and there's a config
file /etc/default/useradd, then it's not passed to useradd in order to let
him choose.
Signed-off-by: Thomas Osterried <ax25@x-berg.in-berlin.de>
|
| |
|
|
|
|
|
| |
Their value has always been dubious and git doesn't support them so let's
get rid of the clutter.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
|
| | |
|
| |
|
|
| |
process group leader, setsid() fails.
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
login -f -h <protocol> <as_user>
did not work.
login -h <protocol> -f <as_user>
did work. Imho, OpenSUSE meessed up login's getopt()...
On the other hand, we have no problems to switch the position
of arguments - as long as there's not another linux demanding
another position ;)
|
| |
|
|
|
| |
In a world that is caught between ISO 8859-1, ISO 8859-15 and UTF-8
non-ASCII characters are a guaranteed mess, still.
|
| |
|
|
|
|
| |
Glibc as old as 2.2 provides working headers under the right name and
if not libax25 would install headers under the right name so this has
become redundant and ugly.
|
| |
|
|
|
|
|
|
| |
Various socket syscalls receive a length argument that should be a
socklen_t rsp. a ptr to a socklen_t but instead int rsp. ptr to int were
being passed. While in theory this was a bug it's harmless as dangerously
large values would not be used but the issue manifested itself in a
significant number of compilation warnings.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Thanks to using the antiquated time(2) call the microsecond field of the
struct utmp being manipulated was never initialized resulting in random
values. On some 64-bit systems this also manifested itself in below
warnings.
axspawn.c: In function ‘cleanup’:
axspawn.c:1017: warning: passing argument 1 of ‘time’ from incompatible pointer type
/usr/include/time.h:186: note: expected ‘time_t *’ but argument is of type ‘int32_t *’
[...]
axspawn.c: In function ‘main’:
[...]
axspawn.c:1693: warning: passing argument 1 of ‘time’ from incompatible pointer type
/usr/include/time.h:186: note: expected ‘time_t *’ but argument is of type ‘int32_t *’
Fixed by changing it to the recommended initialization sequenze based on
gettimeofday().
|
| | |
|
| |
|
|
| |
sizeof(ut.ut_line) with size = 12, wheras ut_id is 4.
|
| |
|
|
| |
never closed.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
on those systems, things are not as you expect. i.e., /bin/login
is a shell script and implements boot-specific actions only
(login without root password, etc..). The assumption that
/bin/login -f username causes user as uid > 0 to login
is wrong; user may gain root rights.
axspawn now implements partially (thus, in the important
points) what login assures.
axspawn should be involked with the option --embeded when
called from ax25d.
|
| |
|
|
|
| |
user has now also the choice to use the standard unix
passwd/shadow authentication
|
| |
|
|
|
|
|
|
|
|
| |
then .profile is copied, and without deletion the contense of
ax25.profile is overwritten. to .profile. This lead to garbage.
- Some useradd implementations refuse to add the user if his
homedir already exists. But we need the function which builds the
subtree /home/hams/dl9.../, because useradd does not make the
missing subdirs by itself (instead, it's a severe error condition
for useradd)
|
| | |
|
| |
|
|
| |
pointless comments.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
intr = ^C as stty setting
|
| |
|
|
|
| |
Documented in axspawn.8
Thanks to Christoph <dk2crn> for contribution.
|
| |
|
|
|
|
|
|
|
| |
- new password methods (read pwcheck for more info)
- mode for homedirectory: new config option secure_home.
if not set, new homedirectories become drwxr-xr-x
if set, new homedirectories become drwx-----x. other=x, because
otherwise uses would be iritating that for e.g. their public_html
is not accessable..
|
| |
|
|
|
|
|
|
|
| |
-EAGAIN, when buffers get empty, which leaded to data loss)
- fixed bug when '\r\n' was splited into two reads (trailing \r in the
first read, starting \n in the second read) -> resulted in an empty line
- better FD_SET(&fdread) handling
- added //COMP hufman compression
- could make use of the useradd(8) program for generating new accounts
|
| | |
|
| | |
|
| | |
|
| |
|
