The following reply was made to PR os-linux/3897; it has been noted by GNATS.
From: Gaven Cohen <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: apache-bugdb@apache.org, [EMAIL PROTECTED] Subject: Re: os-linux/3897: 1.3.4 server starts, seemingly normally... dies, no children, no errors not the case with 1.2.0 Date: Sun, 14 Feb 1999 13:41:31 +0200 (GMT+0200) Hi Marc, Thanks for the quick reply. On 13 Feb 1999 [EMAIL PROTECTED] wrote: > What does strace-ing httpd while starting it do? Alright, being more awake I decided to do an strace -f this time, which shows a segfault. I added in a sleep and gdb'd the child process (the detached parent) and it seems to be dying during the return from set_signals() called from standalone_main(): (gdb) info break Num Type Disp Enb Address What 1 breakpoint keep y 0x0807411e in set_signals at http_main.c:2794 2 breakpoint keep y 0x0807411e in set_signals at http_main.c:2797 3 breakpoint keep y 0x08075ec7 in standalone_main at http_main.c:4274 4 breakpoint keep y 0x08075ecc in standalone_main at http_main.c:4275 (gdb) continue Continuing. Breakpoint 3, standalone_main (argc=1, argv=0xbffffbe4) at http_main.c:4274 4274 set_signals(); (gdb) Continuing. Breakpoint 1, set_signals () at http_main.c:2797 2797 } (gdb) Continuing. Program received signal SIGSEGV, Segmentation fault. 0x0 in ?? () (its not reaching line 4275, the first line after returning) something i forgot to mention previously is that 1.2.0 was compiled libc5, and 1.3.4 glibc. could this be a library problem? Thanks, Gaven ------------------------------------------------------------------------ Gaven Cohen aka Kinslayer http://wastelands.net/ [EMAIL PROTECTED] hkp://keys.pgp.com/0x2042AD07 freelance sysadmin/programmer linux, fantasy, female enthusiast ------------------------------------------------------------------------ RSA/1024: fingerprint 79 BC B3 3F E2 05 71 4B F7 C8 B2 45 EF 70 55 D1
