Celalettin KILINÇ wrote:
Hi,
I have openldap 2.4.36 installed using mdb backend on Solaris 10. I am using this database for a while and suddenly I started getting core dumps while trying to add any data to database. Disk, memory usages seem fine. Database size is not exceeded, it is almost 1/10 full. What may be the reason for this? Please find the other information about the issue as below:
Loaded symbols for /platform/SUNW,Netra-T2000/lib/sparcv9/libc_psr.so.1 Reading symbols from /lib/sparcv9/ld.so.1...(no debugging symbols found)...done. Loaded symbols for /lib/sparcv9/ld.so.1 Core was generated by `./slapd -d5'. Program terminated with signal 10, Bus error. [New process 1335262 ] [New process 90078 ] [New process 155614 ] [New process 221150 ] [New process 286686 ] [New process 352222 ] [New process 417758 ] [New process 483294 ] [New process 548830 ] [New process 614366 ] [New process 679902 ] [New process 745438 ] [New process 810974 ] [New process 876510 ] [New process 942046 ] [New process 1007582 ] [New process 1073118 ] [New process 1138654 ] [New process 1204190 ] [New process 1269726 ] #0 0x0000000100159c3c in mdb_page_search_root () (gdb) bt #0 0x0000000100159c3c in mdb_page_search_root () #1 0x000000010015a2d0 in mdb_page_search () #2 0x000000010015c984 in mdb_cursor_first () #3 0x000000010015d93c in mdb_cursor_get () #4 0x000000010015204c in mdb_page_alloc () #5 0x0000000100152b6c in mdb_page_touch () #6 0x000000010015a264 in mdb_page_search () #7 0x000000010015dba0 in mdb_cursor_touch () #8 0x000000010015e314 in mdb_cursor_put () #9 0x000000010018f25c in mdb_dn2id_add () #10 0x0000000100184e94 in mdb_add () #11 0x0000000100060970 in fe_op_add () #12 0x00000001000601fc in do_add () #13 0x000000010005128c in connection_operation () #14 0x0000000100051a4c in connection_read_thread () #15 0xffffffff7f4164bc in ldap_int_thread_pool_wrapper () from /some/dir/lib/libldap_r-2.4.so.2 #16 0xffffffff7ced7374 in _lwp_start () from /lib/64/libc.so.1 #17 0xffffffff7ced7374 in _lwp_start () from /lib/64/libc.so.1 Backtrace stopped: previous frame identical to this frame (corrupt stack?)
Doesn't look like any crashes we've seen before, but your best bet would be to slapcat the DB and upgrade to a current version. (2.4.39 was released more than 6 months ago, 2.4.40 release candidate is in testing now.) Ideally test against the 2.4.40rc and see if the problem remains.
To be precise - slapcat your DB with your current version, then install the new version and load a new copy of the DB with the new slapadd.