crash during joining node SST phase

[Seppo Jaakola]

It has been reported, that a new node join operation can fail in snapshot transfer phase. More information from the issue will be added soon.

[Teemu Ollakka (teemu-ollakka)]

Crash seems to follow immediately the following log line(s) on joiner: "100108 16:30:16 [Note] WSREP: SST kill local trx: 2888".

Some stack traces:

*** glibc detected *** /home/teemu/galera/mysql/mysql/libexec/mysqld: corrupted double-linked list: 0x954dee88 ***
======= Backtrace: =========
/lib/tls/i686/cmov/libc.so.6[0xb7c6cea9]
/lib/tls/i686/cmov/libc.so.6[0xb7c6e077]
/lib/tls/i686/cmov/libc.so.6(cfree+0x96)[0xb7c6e456]
/home/teemu/galera/mysql/mysql/libexec/mysqld(_Z29one_thread_per_connection_endP3THDb+0x5b)[0x81f883b]
/home/teemu/galera/mysql/mysql/libexec/mysqld(handle_one_connection+0xfe)[0x8200f5e]
/lib/tls/i686/cmov/libpthread.so.0[0xb7dd750f]
/lib/tls/i686/cmov/libc.so.6(clone+0x5e)[0xb7cde7ee]
======= Memory map: ========
[... skip ...]
100108 16:30:16 - mysqld got signal 6 ;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
We will try our best to scrape up some info that will hopefully help diagnose
the problem, but since we have already crashed, something is definitely wrong
and this may fail.

key_buffer_size=8384512
read_buffer_size=131072
max_used_connections=9
max_threads=151
threads_connected=3
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 337741 K
bytes of memory
Hope that's ok; if not, decrease some variables in the equation.
thd: 0x954e81a0
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 0x95bee3b8 thread_stack 0x30000
/home/teemu/galera/mysql/mysql/libexec/mysqld(my_print_stacktrace+0x21)[0x84df3d1]
/home/teemu/galera/mysql/mysql/libexec/mysqld(handle_segfault+0x3d9)[0x81f66e9]
[0xb7f79400]
/lib/tls/i686/cmov/libc.so.6(abort+0x188)[0xb7c2a248]
/lib/tls/i686/cmov/libc.so.6[0xb7c6610d]
/lib/tls/i686/cmov/libc.so.6[0xb7c6cea9]
/lib/tls/i686/cmov/libc.so.6[0xb7c6e077]
/lib/tls/i686/cmov/libc.so.6(cfree+0x96)[0xb7c6e456]
/home/teemu/galera/mysql/mysql/libexec/mysqld(_Z29one_thread_per_connection_endP3THDb+0x5b)[0x81f883b]
/home/teemu/galera/mysql/mysql/libexec/mysqld(handle_one_connection+0xfe)[0x8200f5e]
/lib/tls/i686/cmov/libpthread.so.0[0xb7dd750f]
/lib/tls/i686/cmov/libc.so.6(clone+0x5e)[0xb7cde7ee]
Trying to get some variables.
Some pointers may be invalid and cause the dump to abort...
thd->query at (nil) is an invalid pointer
thd->thread_id=2888
thd->killed=KILL_CONNECTION

100113 19:48:20 - mysqld got signal 11 ;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
We will try our best to scrape up some info that will hopefully help diagnose
the problem, but since we have already crashed, something is definitely wrong
and this may fail.

key_buffer_size=8384512
read_buffer_size=131072
max_used_connections=8
max_threads=151
threads_connected=6
It is possible that mysqld could use ...

[Seppo Jaakola (seppo-jaakola)]

Problem was in wsrep_close_client_connections(), where LOCK_thread_count mutex was not held, when closing connections. This caused a crash, if joining node had incoming client connections, before the SST was initiated.