The Hermit Hacker
2000-Jan-26 14:11 UTC
[2.0.6] doesn't run under Solaris 7/sparc, Internal Error
Is the following a known problem:
==============================================================[2000/01/26
09:41:03, 0] lib/fault.c:fault_report(41)
INTERNAL ERROR: Signal 11 in pid 22273 (2.0.6)
Please read the file BUGS.txt in the distribution
[2000/01/26 09:41:03, 0] lib/fault.c:fault_report(43)
==============================================================
This is using the same config file as I used from 2.0.5, that is currently
running quite well on the same machine ... no apparent core files to be
found ...
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
The Hermit Hacker
2000-Jan-27 13:35 UTC
[2.0.6] doesn't run under Solaris 7/sparc, Internal Error
On Wed, 26 Jan 2000, David Collier-Brown wrote:> You wrote: > ==============================================================> [2000/01/26 09:41:03, 0] lib/fault.c:fault_report(41) > INTERNAL ERROR: Signal 11 in pid 22273 (2.0.6) > Please read the file BUGS.txt in the distribution > [2000/01/26 09:41:03, 0] lib/fault.c:fault_report(43) > > Could you start smbd under truss and email me and the list the > last bit?Not sure how much of the last bit you want, but here's a chunk of it, more if you want it...and just in case Ihaven't mentioned, this is a Solaris 7/sparc system ... 23177: putmsg(5, 0xEFFFD9A8, 0xEFFFD99C, 0) = 0 23177: open("/etc/.syslog_door", O_RDONLY) = 9 23177: door_info(9, 0xEFFFD8E0) = 0 23177: getpid() = 23177 [23175] 23177: door_call(9, 0xEFFFD8C8) = 0 23177: close(9) = 0 23177: write(4, " = = = = = = = = = =".., 66) = 66 23177: time() = 948979975 23177: time() = 948979975 23177: fstat(5, 0xEFFFE658) = 0 23177: time() = 948979975 23177: getpid() = 23177 [23175] 23177: putmsg(5, 0xEFFFDD10, 0xEFFFDD04, 0) = 0 23177: open("/etc/.syslog_door", O_RDONLY) = 9 23177: door_info(9, 0xEFFFDC48) = 0 23177: getpid() = 23177 [23175] 23177: door_call(9, 0xEFFFDC30) = 0 23177: close(9) = 0 23177: write(4, " [", 1) = 1 23177: write(4, " 2 0 0 0 / 0 1 / 2 7 0".., 19) = 19 23177: write(4, " , 0", 3) = 3 23177: write(4, " ] ", 2) = 2 23177: write(4, " l i b / u t i l . c", 10) = 10 23177: write(4, " :", 1) = 1 23177: write(4, " s m b _ p a n i c", 9) = 9 23177: write(4, " (", 1) = 1 23177: write(4, " 2 4 5 6", 4) = 4 23177: write(4, " )\n", 2) = 2 23177: fstat(5, 0xEFFFE280) = 0 23177: time() = 948979975 23177: getpid() = 23177 [23175] 23177: putmsg(5, 0xEFFFD938, 0xEFFFD92C, 0) = 0 23177: open("/etc/.syslog_door", O_RDONLY) = 9 23177: door_info(9, 0xEFFFD870) = 0 23177: getpid() = 23177 [23175] 23177: door_call(9, 0xEFFFD858) = 0 23177: close(9) = 0 23177: write(4, " P A N I C : i n t".., 24) = 24 23177: fstat(5, 0xEFFFE6F0) = 0 23177: time() = 948979975 23177: getpid() = 23177 [23175] 23177: putmsg(5, 0xEFFFDDA8, 0xEFFFDD9C, 0) = 0 23177: open("/etc/.syslog_door", O_RDONLY) = 9 23177: door_info(9, 0xEFFFDCE0) = 0 23177: getpid() = 23177 [23175] 23177: door_call(9, 0xEFFFDCC8) = 0 23177: close(9) = 0 23177: sigaction(SIGABRT, 0x00000000, 0xEFFFEE78) = 0 23177: llseek(0, 0, SEEK_CUR) = 0 23177: sigaction(SIGABRT, 0xEFFFED48, 0xEFFFEDC8) = 0 23177: sigprocmask(SIG_UNBLOCK, 0xEFFFEDF8, 0x00000000) = 0 23177: getpid() = 23177 [23175] 23177: kill(23177, SIGABRT) = 0 23177: Received signal #6, SIGABRT [default] 23177: siginfo: SIGABRT pid=23177 uid=0 23175: Received signal #18, SIGCLD, in poll() [caught] 23175: siginfo: SIGCLD CLD_KILLED pid=23177 status=0x0006 23175: poll(0xEFFFC6A8, 1, -1) Err#91 ERESTART 23175: waitid(P_ALL, 0, 0xEFFFC268, WEXITED|WTRAPPED|WNOHANG) = 0 23175: waitid(P_ALL, 0, 0xEFFFC268, WEXITED|WTRAPPED|WNOHANG) Err#10 ECHILD 23175: setcontext(0xEFFFC488) 23177: *** process killed *** 23175: poll(0xEFFFC6A8, 1, -1) (sleeping...) Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
David Collier-Brown
2000-Jan-27 19:44 UTC
[2.0.6] doesn't run under Solaris 7/sparc, Internal Error
"Marc G. Fournier" wrote:> Here she be, the whole truss output ... >And the interesting sequence is: open64("/usr/local/samba/var/locks/STATUS..LCK", O_RDWR) = 9 ... update the lock file, then ... close(9) ... stat64("/usr/local/lib/smb.conf", 0xEFFFED40) = 0 time() = 948979974 23177: getpid() = 23177 [23175] 23177: brk(0x0013E810) = 0 23177: brk(0x0014E810) = 0 23177: brk(0x0014E810) = 0 23177: brk(0x0015E810) = 0 23177: open64("/usr/slocal/samba/lib/lmhosts", O_RDONLY) Err#2 ENOENT ... then some socket operations on (a new) fd 9, then 23177: close(9) = 0 23177: getuid() = 0 [0] 23177: fstat64(4, 0xEFFFECA0) = 0 23177: write(4, " [", 1) = 1 23177: write(4, " 2 0 0 0 / 0 1 / 2 7 0".., 19) = 19 23177: write(4, " , 0", 3) = 3 23177: write(4, " ] ", 2) = 2 23177: write(4, " l i b s m b / c l i e n".., 18) = 18 23177: write(4, " :", 1) = 1 23177: write(4, " a t t e m p t _ n e t b".., 31) = 31 23177: write(4, " (", 1) = 1 23177: write(4, " 3 2 5 8", 4) = 4 23177: write(4, " )\n", 2) = 2 23177: Incurred fault #6, FLTBOUNDS %pc = 0x0006E148 23177: siginfo: SIGSEGV SEGV_MAPERR addr=0x0000000E 23177: Received signal #11, SIGSEGV [caught] 23177: siginfo: SIGSEGV SEGV_MAPERR addr=0x0000000E This snippet make it look like smbd put out an error message mentioning "attempt_netb"... and segv'd. What are the last half-dozen message in the log? One of them should be "attempt_netbios_session_request: %s rejected the session. Error was : %s.\n" This appears once, at line 3257 in libsmb/clientgen.c: File Line 1 clientgen.c 3224 BOOL attempt_netbios_session_request(struct cli_state *cli, char *srchost, char *desthost, struct in_addr *pdest_ip) which is called from: 3 cli_netlogon.c 532 if (!attempt_netbios_session_request(&cli, global_myname, remote_machine, &cli.dest_ip)) { 4 password.c 998 if (!attempt_netbios_session_request(cli, global_myname, desthost, &dest_ip)) 5 password.c 1158 if (!attempt_netbios_session_request(pcli, global_myname, remote_machine, &dest_ip)) { ----- Ok, that's where it died, smack in the middle of a compound if, and the error message in the log should tell us how to reproduce the failure and the bug, plus may help us avoid it... --dave