Bug#824191: libc6: libpthread SEGFAULT on Xeon CPU E3-1230 v5
Package: libc6:amd64
Version: 2.19-18+deb8
Severity: important
The OpenNebula daemon (oned) segfaults reproducible at start-up.
/var/log/syslog :
...
kernel: traps: oned[7383] general protection ip:7fa9ab07d2c8
sp:7ffd988753a8 error:0 in libpthread-2.19.so[7fa9ab06c000+18000]
...
GDB backtrace
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffff7fd9740 (LWP 9344)]
__lll_unlock_elision (lock=0x8d1f08, private=0) at
../nptl/sysdeps/unix/sysv/linux/x86/elision-unlock.c:29
29 ../nptl/sysdeps/unix/sysv/linux/x86/elision-unlock.c: No such file or
directory.
(gdb) bt
#0 __lll_unlock_elision (lock=0x8d1f08, private=0) at
../nptl/sysdeps/unix/sysv/linux/x86/elision-unlock.c:29
#1 0x00000000004bed5f in PoolObjectSQL::~PoolObjectSQL (this=0x8d1e20,
__in_chrg=<optimized out>)
at include/PoolObjectSQL.h:130
#2 0x00000000005569ea in Zone::~Zone (this=0x8d1e20,
__in_chrg=<optimized out>) at src/zone/Zone.cc:50
#3 0x0000000000556a44 in Zone::~Zone (this=0x8d1e20,
__in_chrg=<optimized out>) at src/zone/Zone.cc:53
#4 0x00000000004fa95c in PoolSQL::allocate (this=0x8ccc10,
objsql=0x8d1e20, error_str="")
at src/pool/PoolSQL.cc:131
#5 0x00000000005562d2 in ZonePool::allocate (this=0x8ccc10,
zone_template=0x8cdce0, oid=0x7fffffffd78c,
error_str="") at src/zone/ZonePool.cc:143
#6 0x0000000000555fae in ZonePool::ZonePool (this=0x8ccc10,
db=0x8aa8b0, is_federation_slave=false)
at src/zone/ZonePool.cc:76
#7 0x0000000000414fc8 in Nebula::start (this=0x858200
<Nebula::instance()::nebulad>, bootstrap_only=false)
at src/nebula/Nebula.cc:492
#8 0x0000000000411580 in oned_main () at src/nebula/oned.cc:85
#9 0x00000000004118ab in main (argc=1, argv=0x7fffffffe6f8) at
src/nebula/oned.cc:218
It seems to be hardware related as the same setup works on another machine.
I already installed intel-microcode which does not solve the problem.
# dmesg | grep micro
[ 1.969950] microcode: CPU0 sig=0x506e3, pf=0x2, revision=0x74
[ 1.976579] microcode: CPU1 sig=0x506e3, pf=0x2, revision=0x74
[ 1.983236] microcode: CPU2 sig=0x506e3, pf=0x2, revision=0x74
[ 1.989875] microcode: CPU3 sig=0x506e3, pf=0x2, revision=0x74
[ 1.996516] microcode: CPU4 sig=0x506e3, pf=0x2, revision=0x74
[ 2.003154] microcode: CPU5 sig=0x506e3, pf=0x2, revision=0x74
[ 2.009793] microcode: CPU6 sig=0x506e3, pf=0x2, revision=0x74
[ 2.016430] microcode: CPU7 sig=0x506e3, pf=0x2, revision=0x74
[ 2.023118] microcode: Microcode Update Driver: v2.00
<tigran@aivazian.fsnet.co.uk>, Peter Oruba
Maybe it is related to [1] and TSX should be disabled for this CPU.
References
[1]
http://www.mail-archive.com/debian-bugs-dist%40lists.debian.org/msg1360182.html
Regards,
Benjamin
Reply to: