zabbix 4.0.3 history syncer process is 100 % busy

[Gocha]

Hello!

Need help!

We've upgraded our Zabbix from 3.0.10 and finally to 4.0.3 with MySQL DB (Zabbix DB is partitioned) on the same phisycal server.
All looks fine, except periodically every hour or so zabbix history syncer process is 100 % busy.
Furthermore, at this period last date for all elements isn't updated in Zabbix WEB.
Log file for zabbix server contains information about slow querries during this period.
Our old Zabbix 3.0.10 is also running and it doesn't have such symptoms!

Where could be the problem with Zabbix 4.0.3?

[Gocha]

Server's 32G RAM.
_______________

Zabbix configuration:

StartPollers=250

StartPollersUnreachable=60

StartTrappers=10

StartPingers=30

StartDiscoverers=70

SNMPTrapperFile=/var/log/snmptt/snmptt.log

StartSNMPTrapper=1

MaxHousekeeperDelete=1500

CacheSize=1G

StartDBSyncers=8

HistoryCacheSize=192M

HistoryIndexCacheSize=192M

TrendCacheSize=80M

ValueCacheSize=640M

Timeout=10

AlertScriptsPath=/usr/lib/zabbix/alertscripts

ExternalScripts=/usr/lib/zabbix/externalscripts

LogSlowQueries=3000

_________________

MySQL configuration:

[mysqld]


max_connections = 1000
connect_timeout = 5
wait_timeout = 600
max_allowed_packet = 160M
thread_cache_size = 128
sort_buffer_size = 4M
bulk_insert_buffer_size = 16M
tmp_table_size = 32M
max_heap_table_size = 32M

myisam_recover_options = BACKUP
key_buffer_size = 128M
table_open_cache = 2000
myisam_sort_buffer_size = 512M
concurrent_insert = 2
read_buffer_size = 2M
read_rnd_buffer_size = 1M

query_cache_limit = 128K
query_cache_size = 64M

slow_query_log = 1
slow_query_log_file = /var/log/mysql/mariadb-slow.log
long_query_time = 10
log_slow_verbosity = query_plan

log_bin = /var/lib/mysql/binlogs/mariadb-bin
log_bin_index = /var/lib/mysql/binlogs/mariadb-bin.index
expire_logs_days = 10
max_binlog_size = 100M

gtid_strict_mode = TRUE

default_storage_engine = InnoDB

innodb_log_file_size = 256M
innodb_buffer_pool_size = 16G
innodb_log_buffer_size = 16M
innodb_file_per_table = 1
innodb_open_files = 2048
innodb_io_capacity = 300
innodb_flush_method = O_DIRECT
innodb_file_format = Barracuda
innodb_flush_log_at_trx_commit = 2
innodb_stats_on_metadata = OFF
innodb_undo_tablespaces = 2
innodb_undo_logs = 35

ignore_db_dirs = lost+found
ignore_db_dirs = binlogs

[mysqldump]
quick
quote-names
max_allowed_packet = 160M

[mysql]


[isamchk]
key_buffer = 16M

[dimir]

Could you try decreasing number of DBSyncers to default?

StartDBSyncers=4

[Gocha]

Now StartDBSyncers=4.
Please, look at the attachments.

To my mind, changing DBSyncers to default helps, doesn't it?

[dimir]

Yes, it looks much better to me.

[Gocha]

Hope, it helps. For the last two days it's all right!
Thank you very much!