Nodes are randomly crashing

Hello,

we have cluster with 3 nodes and during last days we did upgrade from latest 10.3 to 10.4.6 After this upgrade we have problem with randomly crashing nodes:

{{Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: cluster conflict due to high priority abort for threads:
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: Winning thread:
Jul 11 14:00:04 SQL1 mysqld: THD: 668932, mode: toi, state: exec, conflict: aborted, seqno: 4703314
Jul 11 14:00:04 SQL1 mysqld: SQL: TRUNCATE [XXXXXXXXX]
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: Victim thread:
Jul 11 14:00:04 SQL1 mysqld: THD: 668959, mode: local, state: exec, conflict: executing, seqno: -1
Jul 11 14:00:04 SQL1 mysqld: SQL: INSERT INTO [XXXXXXXXX]
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: context: /home/buildbot/buildbot/build/mariadb-10.4.6/storage/innobase/handler/ha_innodb.cc:18658
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: MDL conflict db=[XXXXXXXXX] table=[XXXXXXXXX] ticket=3 solved by abort
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: cluster conflict due to high priority abort for threads:
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: Winning thread:
Jul 11 14:00:04 SQL1 mysqld: THD: 668932, mode: toi, state: exec, conflict: aborted, seqno: 4703314
Jul 11 14:00:04 SQL1 mysqld: SQL: TRUNCATE [XXXXXXXXX]
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: Victim thread:
Jul 11 14:00:04 SQL1 mysqld: THD: 668962, mode: local, state: exec, conflict: executing, seqno: -1
Jul 11 14:00:04 SQL1 mysqld: SQL: INSERT INTO [XXXXXXXXX]
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: context: /home/buildbot/buildbot/build/mariadb-10.4.6/storage/innobase/handler/ha_innodb.cc:18658
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668932 [Note] WSREP: MDL conflict db=[XXXXXXXXX] table=[XXXXXXXXX] ticket=3 solved by abort
Jul 11 14:00:04 SQL1 mysqld: 2019-07-11 14:00:04 668959 [Note] WSREP: MDL conflict db=[XXXXXXXXX] table=[XXXXXXXXX] ticket=3 solved by abort
Jul 11 14:00:04 SQL1 mysqld: mysqld: /home/buildbot/buildbot/build/mariadb-10.4.6/wsrep-lib/src/client_state.cpp:232: int wsrep::client_state::after_statement(): Assertion `transaction_.state() == wsrep::transaction::s_aborted' failed.
Jul 11 14:00:04 SQL1 mysqld: 190711 14:00:04 [ERROR] mysqld got signal 6 ;
Jul 11 14:00:04 SQL1 mysqld: This could be because you hit a bug. It is also possible that this binary
Jul 11 14:00:04 SQL1 mysqld: or one of the libraries it was linked against is corrupt, improperly built,
Jul 11 14:00:04 SQL1 mysqld: or misconfigured. This error can also be caused by malfunctioning hardware.
Jul 11 14:00:04 SQL1 mysqld:
Jul 11 14:00:04 SQL1 mysqld: To report this bug, see https://mariadb.com/kb/en/reporting-bugs
Jul 11 14:00:04 SQL1 mysqld:
Jul 11 14:00:04 SQL1 mysqld: We will try our best to scrape up some info that will hopefully help
Jul 11 14:00:04 SQL1 mysqld: diagnose the problem, but since we have already crashed,
Jul 11 14:00:04 SQL1 mysqld: something is definitely wrong and this may fail.
Jul 11 14:00:04 SQL1 mysqld:
Jul 11 14:00:04 SQL1 mysqld: Server version: 10.4.6-MariaDB-1:10.4.6+maria~stretch-log
Jul 11 14:00:04 SQL1 mysqld: key_buffer_size=67108864
Jul 11 14:00:04 SQL1 mysqld: read_buffer_size=1048576
Jul 11 14:00:04 SQL1 mysqld: max_used_connections=198
Jul 11 14:00:04 SQL1 mysqld: max_threads=5002
Jul 11 14:00:04 SQL1 mysqld: thread_count=223
Jul 11 14:00:04 SQL1 mysqld: It is possible that mysqld could use up to
Jul 11 14:00:04 SQL1 mysqld: key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 66774957 K bytes of memory
Jul 11 14:00:04 SQL1 mysqld: Hope that's ok; if not, decrease some variables in the equation.
Jul 11 14:00:04 SQL1 mysqld:
Jul 11 14:00:04 SQL1 mysqld: Thread pointer: 0x7fbfec0009a8
Jul 11 14:00:04 SQL1 mysqld: Attempting backtrace. You can use the following information to find out
Jul 11 14:00:04 SQL1 mysqld: where mysqld died. If you see no messages after this, something went
Jul 11 14:00:04 SQL1 mysqld: terribly wrong...
Jul 11 14:00:04 SQL1 mysqld: mysqld: /home/buildbot/buildbot/build/mariadb-10.4.6/wsrep-lib/src/client_state.cpp:232: int wsrep::client_state::after_statement(): Assertion `transaction_.state() == wsrep::transaction::s_aborted' failed.
Jul 11 14:00:07 SQL1 mysqld_safe: Number of processes running now: 0
Jul 11 14:00:07 SQL1 mysqld_safe: WSREP: not restarting wsrep node automatically
Jul 11 14:00:07 SQL1 mysqld_safe: mysqld from pid file /var/run/mysqld/mysqld.pid ended
}}