Bug #83798 handle_fatal_signal (sig=11) in my_valid_mbcharlen_utf8

Description:
Version: '5.6.34-debug'  socket: '/sda/MS081116-mysql-5.6.34-linux-x86_64-debug/socket.sock'  port: 11652  MySQL Community Server (GPL)
2016-11-14 15:00:01 15681 [Note] Event Scheduler: scheduler thread started with id 2
04:00:02 UTC - mysqld got signal 11 ;

(gdb) bt
#0  0x00007fdab7b37741 in __pthread_kill (threadid=<optimized out>, signo=11) at ../nptl/sysdeps/unix/sysv/linux/pthread_kill.c:61
#1  0x0000000000a9c852 in my_write_core (sig=11) at /git/mysql-server_dbg/mysys/stacktrace.c:422
#2  0x000000000072c554 in handle_fatal_signal (sig=11) at /git/mysql-server_dbg/sql/signal_handler.cc:230
#3  <signal handler called>
#4  0x0000000000e208e9 in my_valid_mbcharlen_utf8 (cs=0x17238a0 <my_charset_utf8_bin>, s=0x0, e=0x4 <Address 0x4 out of bounds>)
    at /git/mysql-server_dbg/strings/ctype-utf8.c:5974
#5  0x0000000000e20958 in my_well_formed_len_utf8 (cs=0x17238a0 <my_charset_utf8_bin>, b=0x0, e=0x4 <Address 0x4 out of bounds>, pos=64, error=0x7fdab80e9e48)
    at /git/mysql-server_dbg/strings/ctype-utf8.c:6035
#6  0x000000000083812f in well_formed_copy_nchars (to_cs=0x17238a0 <my_charset_utf8_bin>, to=0x7fda89a42c12 "test", ' ' <repeats 188 times>, "e_43    "..., to_length=192,
    from_cs=0x147c3a0 <my_charset_bin>, from=0x0, from_length=4, nchars=64, well_formed_error_pos=0x7fdab80e9ef0, cannot_convert_error_pos=0x7fdab80e9ee8,
    from_end_pos=0x7fdab80e9ee0) at /git/mysql-server_dbg/sql/sql_string.cc:1015
#7  0x0000000000909131 in Field_string::store (this=0x7fda89aa8728, from=0x0, length=4, cs=0x147c3a0 <my_charset_bin>) at /git/mysql-server_dbg/sql/field.cc:6658
#8  0x00000000009e1cff in Event_db_repository::find_named_event (this=0x7fda9bbfe350, db=..., name=..., table=0x7fda89a42200)
    at /git/mysql-server_dbg/sql/event_db_repository.cc:974
#9  0x00000000009e260a in Event_db_repository::update_timing_fields_for_event (this=0x7fda9bbfe350, thd=0x7fda82c74000, event_db_name=..., event_name=..., last_executed=0,
    status=0) at /git/mysql-server_dbg/sql/event_db_repository.cc:1142
#10 0x00000000009e5cba in Event_queue::get_top_for_execution_if_time (this=0x7fdab4022de0, thd=0x7fda82c74000, event_name=0x7fdab80ead10)
    at /git/mysql-server_dbg/sql/event_queue.cc:672
#11 0x00000000009e7874 in Event_scheduler::run (this=0x7fda9afa1a80, thd=0x7fda82c74000) at /git/mysql-server_dbg/sql/event_scheduler.cc:499
#12 0x00000000009e6ea7 in event_scheduler_thread (arg=0x7fda82c19070) at /git/mysql-server_dbg/sql/event_scheduler.cc:242
#13 0x0000000000ae8ac9 in pfs_spawn_thread (arg=0x7fda82cfc020) at /git/mysql-server_dbg/storage/perfschema/pfs.cc:1860
#14 0x00007fdab7b32dc5 in start_thread (arg=0x7fdab80eb700) at pthread_create.c:308
#15 0x00007fdab6197ced in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:113

How to repeat:
# mysqld options required for replay:  --maximum-max_heap_table_size=1M
DROP DATABASE test;CREATE DATABASE test;USE test;
SET DEBUG='+d,fts_union_limit_off';
SET GLOBAL debug='+d,simulate_out_of_memory';
create event e_43 on schedule every 1 second do set @a=5;
set global event_scheduler=on;

Hello Roel,

Thank you for the report and test case.
Observed that 5.6.34/5.7.16 debug builds are affected.

Thanks,
Umesh

Posted by developer:
 
SET GLOBAL debug='+d,simulate_out_of_memory';
will have *every* call to my_malloc fail, which is not reasonable.
Writing code to handle such a scenario is not productive.