Hiya,
I don’t know if anyone else has seen this issue but in the last couple of days, after adding more storage and as a consequence, getting significantly higher throughput from ALICE, my FSTs are seg faulting every 2-3 hours with:
Aug 19 17:30:05 epgsr6 kernel: xrootd[14136]: segfault at 0 ip 00000037b0a8987b sp 00007fc0943e0918 error 6 in libc-2.12.so[37b0a00000+18b000]
I’m not certain, but neither memory or processor seems to be stretched when it fails and I think I have nofile set correctly:
#
-
hard nofile 65500 -
soft nofile 65500
Am I missing something obvious? Any help would be greatly appreciated!
Many Thanks,
Mark Slater, Birmingham University
UPDATE:
I should add, packages running are:
eos-apmon.x86_64 1.1.8-1.el6 @eos-dependencies
eos-client.x86_64 4.2.29-1.el6 @eos
eos-folly.x86_64 2017.09.18.00-4.el6 @eos-dependencies
eos-server.x86_64 4.2.29-1.el6 @eos
libmicrohttpd.x86_64 0.9.38-eos.yves.slc6 @eos-dependencies
protobuf3.x86_64 3.3.1-1.el6 @eos-dependencies
tokenauthz.x86_64 1.2.0-1 @eos-dependencies
xrootd.x86_64 1:4.8.4-1.el6 @epel
xrootd-alicetokenacc.x86_64 1.3.1-1 @eos-dependencies
xrootd-client.x86_64 1:4.8.4-1.el6 @epel
xrootd-client-devel.x86_64 1:4.8.4-1.el6 @epel
xrootd-client-libs.x86_64 1:4.8.4-1.el6 @epel
xrootd-devel.x86_64 1:4.8.4-1.el6 @epel
xrootd-libs.x86_64 1:4.8.4-1.el6 @epel
xrootd-private-devel.x86_64 1:4.8.4-1.el6 @epel
xrootd-selinux.noarch 1:4.8.4-1.el6 @epel
xrootd-server.x86_64 1:4.8.4-1.el6 @epel
xrootd-server-devel.x86_64 1:4.8.4-1.el6 @epel
xrootd-server-libs.x86_64 1:4.8.4-1.el6 @epel
on SL6.9.
Thanks again!
Mark