[Apologies, I may have sent this before from the wrong mail-box]
Hello I am running the Blastwave packaging of Freeradius 1.0.1 on Solaris 8, sparc. This interfaces with MySQL version 4.0.18-log, built from source. The RADIUS server inexplicably terminates with no core dump or logged messages The tail of a truss log shows bash-2.05# grep SEGV_MAPERR truss.log | wc -l 1766 bash-2.05# tail -20 truss.log 10471/6: 978.8120 Incurred fault #6, FLTBOUNDS %pc = 0xFED714F4 10471/6: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000003 10471/6: 978.8121 Received signal #11, SIGSEGV [caught] 10471/6: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000003 10471/6: 978.8123 sigprocmask(SIG_SETMASK, 0xFF17CFB8, 0x00000000) = 0 10471/6: 978.8124 sigprocmask(SIG_SETMASK, 0xFF188CE0, 0x00000000) = 0 10471/6: 978.8125 setcontext(0xFE702CD0) 10471/6: 978.8127 Incurred fault #6, FLTBOUNDS %pc = 0xFED714F4 10471/6: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000003 10471/6: 978.8128 Received signal #11, SIGSEGV [caught] 10471/6: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000003 10471/6: 978.8130 sigprocmask(SIG_SETMASK, 0xFF17CFB8, 0x00000000) = 0 10471/6: 978.8131 sigprocmask(SIG_SETMASK, 0xFF188CE0, 0x00000000) = 0 10471/6: 978.8132 setcontext(0xFE702CD0) 10471/6: 978.8134 Incurred fault #6, FLTBOUNDS %pc = 0xFED714F4 10471/6: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000003 10471/6: 978.8135 Received signal #11, SIGSEGV [caught] 10471/6: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000003 10471/6: 978.8137 sigprocmask(SIG_SETMASK, 0xFF17CFB8, 0x00000000) = 0 10471/1: 978.8140 _exit(1) It got into a SEGV_MAPERR 1766 times while I had it under truss... This error is superficially similar to the one reported by Terry J Fike Jr in July 2004, but here the server exits voluntarily. Has this problem been encountered before? What further investigation should I do? Thank you, Barry Barry A Byrne Network Applications and Public Services Cable & Wireless - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html