Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Incomplete
-
10.6.12
-
Server version: 10.6.12-MariaDB-0ubuntu0.22.04.1 source revision
VM on hyperv machine with 16GB ram and 4 core
Galera cluster
Description
Hello,
I've got 2nd time issue regarding MariaDB killed a service which made me outage.
First time I've doubled the memory to 16GB (Swap turned off) but seems like not enough even my NeXTcloud server is not heavily used.
Some of my config part:
/etc/mysql/mariadb.conf.d/50-server.cnf |
[server]
|
[mysqld]
|
pid-file = /run/mysqld/mysqld.pid
|
basedir = /usr
|
max_allowed_packet = 256M
|
expire_logs_days = 10 |
character-set-server = utf8mb4
|
collation-server = utf8mb4_general_ci
|
[embedded]
|
[mariadb]
|
[mariadb-10.6] |
[galera]
|
wsrep_on=ON
|
wsrep_provider=/usr/lib/galera/libgalera_smm.so
|
wsrep_cluster_address="gcomm://1.1.1.1,1.1.1.2" |
binlog_format=row
|
transaction_isolation = READ-COMMITTED
|
default_storage_engine=InnoDB
|
innodb_autoinc_lock_mode=2 |
bind-address=0.0.0.0 |
wsrep_cluster_name="MariaDB_Nextcloud_Cluster" |
wsrep_node_name='DB1' |
wsrep_node_address="1.1.1.1" |
innodb_buffer_pool_size = 1G
|
innodb_log_file_size = 256M
|
read_rnd_buffer_size = 4M
|
sort_buffer_size = 4M
|
max_allowed_packet = 1G
|
This is the issue in the logs:
Jan 22 13:30:49 servername mariadbd[1068]: 2024-01-22 13:30:49 514463 [Warning] Aborted connection 514463 to db: 'unconnected' user: 'haproxy' host: 'servername' (Got an error writing communication packets |
)
|
Jan 22 13:30:50 servername mariadbd[1068]: terminate called after throwing an instance of 'boost::wrapexcept<std::system_error>' |
Jan 22 13:30:51 servername mariadbd[1068]: what(): remote_endpoint: Transport endpoint is not connected |
Jan 22 13:30:51 servername mariadbd[1068]: 240122 13:30:51 [ERROR] mysqld got signal 6 ; |
Jan 22 13:30:51 servername mariadbd[1068]: This could be because you hit a bug. It is also possible that this binary |
Jan 22 13:30:51 servername mariadbd[1068]: or one of the libraries it was linked against is corrupt, improperly built, |
Jan 22 13:30:51 servername mariadbd[1068]: or misconfigured. This error can also be caused by malfunctioning hardware. |
Jan 22 13:30:51 servername mariadbd[1068]: To report this bug, see https://mariadb.com/kb/en/reporting-bugs |
Jan 22 13:30:51 servername mariadbd[1068]: We will try our best to scrape up some info that will hopefully help |
Jan 22 13:30:51 servername mariadbd[1068]: diagnose the problem, but since we have already crashed, |
Jan 22 13:30:51 servername mariadbd[1068]: something is definitely wrong and this may fail. |
Jan 22 13:30:51 servername mariadbd[1068]: Server version: 10.6.12-MariaDB-0ubuntu0.22.04.1 source revision: |
Jan 22 13:30:51 servername mariadbd[1068]: key_buffer_size=134217728 |
Jan 22 13:30:51 servername mariadbd[1068]: read_buffer_size=131072 |
Jan 22 13:30:51 servername mariadbd[1068]: max_used_connections=152 |
Jan 22 13:30:51 servername mariadbd[1068]: max_threads=153 |
Jan 22 13:30:51 servername mariadbd[1068]: thread_count=153 |
Jan 22 13:30:51 servername mariadbd[1068]: It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 781311 K bytes of memory |
Jan 22 13:30:51 servername mariadbd[1068]: Hope that's ok; if not, decrease some variables in the equation. |
Jan 22 13:30:51 servername mariadbd[1068]: Thread pointer: 0x0 |
Jan 22 13:30:51 servername mariadbd[1068]: Attempting backtrace. You can use the following information to find out |
Jan 22 13:30:51 servername mariadbd[1068]: where mysqld died. If you see no messages after this, something went |
Jan 22 13:30:51 servername mariadbd[1068]: terribly wrong... |
.
|
.
|
.
|
Jan 22 13:37:12 servername kernel: [516744.982529] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=init.scope,mems_allowed=0,global_oom,task_memcg=/system.slice/php8.2-fpm.service,task=php-fpm8.2,pid=962,uid=33 |
Jan 22 13:37:12 servername kernel: [516744.982547] Out of memory: Killed process 962 (php-fpm8.2) total-vm:680576kB, anon-rss:240744kB, file-rss:3588kB, shmem-rss:101428kB, UID:33 pgtables:912kB oom_score_adj:0 |
Jan 22 13:37:12 servername kernel: [516745.023948] systemd[1]: php8.2-fpm.service: A process of this unit has been killed by the OOM killer. |
.
|
.
|
.
|
Jan 22 13:37:13 servername systemd[1]: php8.2-fpm.service: Failed with result 'oom-kill'. |
Jan 22 13:37:13 servername systemd[1]: php8.2-fpm.service: Consumed 4h 17min 42.625s CPU time. |