I am testing a setup of samba 3.0(beta 2) PDC using OpenLDAP 2.1.21 as the backend. The samba part seems to be functioning alright and even wbinfo -a completes normally as expected. But there seems to be some problem in the libnss_winbind.so : I setup the pam.d/login to use pam_winbind and nsswitch.conf to use libnss_winbind. When I tried to login with the form of "login Domain\\user", pam_winbind completes as expected but libnss_winbind caused the winbindd panic. The panic code is as below : [2003/09/24 14:43:52, 0] lib/fault.c:fault_report(36) =============================================================== [2003/09/24 14:43:52, 0] lib/fault.c:fault_report(37) INTERNAL ERROR: Signal 11 in pid 1280 (3.0.0beta2-0.nobse.1 for Debian) Please read the appendix Bugs of the Samba HOWTO collection [2003/09/24 14:43:52, 0] lib/fault.c:fault_report(39) =============================================================== [2003/09/24 14:43:52, 0] lib/util.c:smb_panic(1452) smb_panic(): calling panic action [/usr/share/samba/panic-action 1280] [2003/09/24 14:43:52, 0] lib/util.c:smb_panic(1460) smb_panic(): action returned status 0 [2003/09/24 14:43:52, 0] lib/util.c:smb_panic(1462) PANIC: internal error [2003/09/24 14:43:52, 0] lib/util.c:smb_panic(1469) BACKTRACE: 12 stack frames: #0 /usr/sbin/winbindd(smb_panic+0x16b) [0x80abd43] #1 /usr/sbin/winbindd [0x809cbb2] #2 /usr/sbin/winbindd [0x809cc05] #3 /lib/libc.so.6 [0x402276b8] #4 /usr/sbin/winbindd(strequal+0x2d) [0x80a48cd] #5 /usr/sbin/winbindd(winbindd_getpwnam+0xba) [0x806cf2a] #6 /usr/sbin/winbindd(ber_scanf+0x69c) [0x806bb8c] #7 /usr/sbin/winbindd(winbind_process_packet+0x20) [0x806be40] #8 /usr/sbin/winbindd [0x806c692] #9 /usr/sbin/winbindd(main+0x493) [0x806cbf3] #10 /lib/libc.so.6(__libc_start_main+0xbb) [0x4021714f] #11 /usr/sbin/winbindd(fsetxattr+0x31) [0x806b571] It seems to be caused by libnss_winbind as if I omit libnss_winbind in nsswitch.conf, the daemon doesn't panic.
please retest against 3.0.1pre2 (due out later today) since beta2 is pretty old at this point and I have a feeling this has been fixed.
reseting target milestone. 3.0.1 has been frozen. WIll have to re-evaluate these.
closing. No feedback in several months. Assuming the segfault has been fixed.
originally reported against 3.0.0beta2. CLeaning out non-production release versions.
sorry for the same, cleaning up the database to prevent unecessary reopens of bugs.
database cleanup