nscd[2320]: segfault at 0 ip 00007f911bd87741 sp 00007f910e202188 error 4 in libc-2.17.so[7f911bc38000+1a3000]
I get frequently the error in syslog files:
nscd[2320]: segfault at 0 ip 00007f911bd87741 sp 00007f910e202188 error 4 in libc-2.17.so
fortunately the systemd re-spawns the nscd immediately. After my quick core dump analysis I'm quite sure that the problem comes from:
glibc-2.17-4.4.1, nscd netgroupcache.c in line 205:
size_t userlen = strlen (nuser) + 1; <------ core dump
because nuser in the triple code is 0 in case of (host,,) entries in NIS! This may causes the core dump.
I found two work arounds to avoid a nscd crash:
1.) switch off netgroup in nscd.conf enable-cache netgroup yes -> no or
2.) in NIS change entries in netgroup (host,,) -> (host,-,<youdomain>
regards Thomas
nscd[2320]: segfault at 0 ip 00007f911bd87741 sp 00007f910e202188 error 4 in libc-2. 17.so[7f911bc38 000+1a3000]
I get frequently the error in syslog files:
nscd[2320]: segfault at 0 ip 00007f911bd87741 sp 00007f910e202188 error 4 in libc-2.17.so
fortunately the systemd re-spawns the nscd immediately. After my quick core dump analysis I'm quite sure that the problem comes from:
glibc-2.17-4.4.1, nscd netgroupcache.c in line 205:
size_t userlen = strlen (nuser) + 1; <------ core dump
because nuser in the triple code is 0 in case of (host,,) entries in NIS! This may causes the core dump.
I found two work arounds to avoid a nscd crash:
1.) switch off netgroup in nscd.conf
enable- cache netgroup yes -> no
or
2.) in NIS change entries in netgroup
(host,,) -> (host,-,<youdomain>
regards Thomas