I'm struggeling with getting core-dumps for some crashes we're seeing
with ATS 3.2.0 (+TS-1392 patch). What we're seeing is this logged in
"dmesg":
[ET_SSL 5][29708]: segfault at 0 ip 00002b7b94f14425 sp
00002b7ba4eadb88 error 4 in libssl.so.1.0.0[2b7b94edf000+53000]
and traffic.out saying:
[Sep 16 17:17:23.316] Manager {0x7f84e91a37e0} ERROR:
[LocalManager::pollMgmtProcessServer] Server Process terminated due to
Sig 11: Segmentation fault
[Sep 16 17:17:23.321] Manager {0x7f84e91a37e0} ERROR: (last system
error 2: No such file or directory)
[Sep 16 17:17:23.321] Manager {0x7f84e91a37e0} ERROR:
[Alarms::signalAlarm] Server Process was reset
[Sep 16 17:17:23.321] Manager {0x7f84e91a37e0} ERROR: (last system
error 2: No such file or directory)
[Sep 16 17:17:24.331] Manager {0x7f84e91a37e0} NOTE:
[LocalManager::startProxy] Launching ts process
[TrafficServer] using root directory '/usr'
[Sep 16 17:17:24.352] Manager {0x7f84e91a37e0} NOTE:
[LocalManager::pollMgmtProcessServer] New process connecting fd '14'
[Sep 16 17:17:24.352] Manager {0x7f84e91a37e0} NOTE:
[Alarms::signalAlarm] Server Process born
[Sep 16 17:17:25.367] {0x2b77790c0d00} STATUS: opened
/var/log/trafficserver/diags.log
[Sep 16 17:17:25.367] {0x2b77790c0d00} NOTE: updated diags config
[Sep 16 17:17:25.371] Server {0x2b77790c0d00} NOTE: cache clustering disabled
[Sep 16 17:17:25.395] Server {0x2b77790c0d00} NOTE: cache clustering disabled
[Sep 16 17:17:25.430] Server {0x2b77790c0d00} ERROR: Cannot insert duplicate!
[Sep 16 17:17:25.431] Server {0x2b77790c0d00} ERROR: Cannot insert duplicate!
[Sep 16 17:17:25.431] Server {0x2b77790c0d00} ERROR: Cannot insert duplicate!
[Sep 16 17:17:25.483] Server {0x2b77790c0d00} NOTE: logging
initialized[15], logging_mode = 3
[Sep 16 17:17:25.536] Server {0x2b77790c0d00} NOTE: traffic server running
[Sep 16 17:17:25.861] Server {0x2b7779e38700} NOTE: cache enabled
I've tried enabling core-dumps the way I got it working earlier, but
see no core dumps.. :
sysctl -w fs.suid_dumpable=1
sysctl -w kernel.core_pattern=/tmp/core.%e.%p
records.config: CONFIG proxy.config.stack_dump_enabled INT 0
/etc/profile: ulimit -c unlimited >/dev/null 2>&1
/etc/sysconfig/init: DAEMON_COREFILE_LIMIT='unlimited'
Does the crash seem familiar to anybody? Or could someone please help
me with other settings needed to enable core dumps on RHEL6.3 so that
I can provide a better bug report ?
-jf
