Hi!
It seems slapd crash when it's hammered with ~30 simultaneous connections in our test environment. Each client connecting to the server run "ldapsearch -YGSSAPI -H ldap://ldap.example.com uid=user". In earlier versions of openldap we noticed crashes in SASL with the same setup, however that doesn't seem to happen anymore. The test server is running openldap 2.4.28 on Ubuntu 12.04. After a while slapd segfault with the following backtrace:
(gdb) bt #0 0x00007f8577401445 in raise () from /lib/x86_64-linux-gnu/libc.so.6 #1 0x00007f8577404bab in abort () from /lib/x86_64-linux-gnu/libc.so.6 #2 0x00007f85773fa10e in ?? () from /lib/x86_64-linux-gnu/libc.so.6 #3 0x00007f85773fa1b2 in __assert_fail () from /lib/x86_64-linux-gnu/libc.so.6 #4 0x00007f8578cf17e0 in slap_listener (sl=0x7f857978b2e0) at ../../../../servers/slapd/daemon.c:1890 #5 0x00007f8578cf17f9 in slap_listener_thread (ctx=<optimized out>, ptr=0x7f857978b2e0) at ../../../../servers/slapd/daemon.c:2092 #6 0x00007f85788579aa in ldap_int_thread_pool_wrapper (xpool=0x7f85797c8c80) at ../../../../libraries/libldap_r/tpool.c:685 #7 0x00007f857778fe9a in start_thread () from /lib/x86_64-linux-gnu/libpthread.so.0 #8 0x00007f85774bd4bd in clone () from /lib/x86_64-linux-gnu/libc.so.6 #9 0x0000000000000000 in ?? ()
For a full bt: https://gist.github.com/112c3bf7a98a5df69907
Let me know if anything else is needed.