|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Re: chroot'ed environment?
Valdis.Kletnieks
vt.eduThu, 20 Apr 1995 03:49:00 -0400
- Messages sorted by: [ date ][ thread ][ subject ][ author ]
- Next message: don
paranoia.com: "Re: chroot'ed environment?"
- Previous message: Brent Chapman: "Re: The Dan Farmer rap"
- In reply to: der Mouse: "Re: chroot'ed environment?"
- Next in thread: oea!owner-bugtraq
oea.xs4all.nl: "Re: chroot'ed environment?"
On Wed, 19 Apr 1995 17:47:13 EDT, you said: > An idea which just occurred to me, not tested at all. If you can > connect() an AF_UNIX SOCK_DGRAM socket (and I'm not sure you can), the > association with its peer might survive a chroot that renders the > original pathname inaccessible. If this is so, it could provide an > answer. Excerpted from 'man 3 openlog' on an AIX 3.2.5 system, but it should be in any BSD4.3 system: #include <syslog.h> int openlog (ID, LogOption, Facility) char *ID; int LogOption, Facility; ... LogOption Specifies a bit field that indicates logging options. The values of LogOption are: ... LOG_NDELAY Opens the connection to the syslogd command immediate- ly, instead of when the first message is logged. This option is useful for programs that need to manage the order in which file descriptors are allocated. ... So if you call openlog(argv[0],LOG_NDELAY,LOG_DAEMON); the LOG_NDELAY will open /dev/log *now* (presumably before you chroot()). And as we all know, open file descriptors are not affected by a chroot().... Valdis Kletnieks Computer Systems Engineer Virginia Tech
- Next message: don
paranoia.com: "Re: chroot'ed environment?"
- Previous message: Brent Chapman: "Re: The Dan Farmer rap"
- In reply to: der Mouse: "Re: chroot'ed environment?"
- Next in thread: oea!owner-bugtraq
oea.xs4all.nl: "Re: chroot'ed environment?"