Strange log message (Got an error reading communication packets)

[InventoR]

Hi.
Could you help me with fix some problem?

We have 5version OTRS Free on CentOS6. When we open otrs dashboard or other ticket we have next message in mysql server:

Code: Select all

2016-06-21T20:19:41.179875Z 319 [Note] Aborted connection 319 to db: 'otrs' user: 'otrs' host: 'localhost' (Got an error reading communication packets)
2016-06-21T20:19:41.180875Z 318 [Note] Aborted connection 318 to db: 'otrs' user: 'otrs' host: 'localhost' (Got an error reading communication packets)
2016-06-21T20:19:52.255524Z 323 [Note] Aborted connection 323 to db: 'otrs' user: 'otrs' host: 'localhost' (Got an error reading communication packets)

Code: Select all

[root@otrs ~]# rpm -qa|grep mysql
php-mysql-5.4.45-54.el6.art.x86_64
mysql-community-client-5.7.13-1.el6.x86_64
mysql-community-server-5.7.13-1.el6.x86_64
mysqlclient16-5.1.59-2.el6.art.x86_64
mysql57-community-release-el6-8.noarch
mysql-community-libs-5.7.13-1.el6.x86_64
mysql-community-common-5.7.13-1.el6.x86_64
[root@otrs ~]# 

my.cnf

Code: Select all

[root@otrs ~]# grep -ve "^;\|^[[:space:]]*$\|^#" /etc/my.cnf
[client]
port            = 3306
socket          = /var/lib/mysql/mysql.sock

[mysqld]
bind-address = 0.0.0.0
port            = 3306
socket          = /var/lib/mysql/mysql.sock
skip-external-locking
key_buffer_size = 128M
max_allowed_packet = 256M
sort_buffer_size = 1M
join_buffer_size=4M
read_buffer_size = 1M
read_rnd_buffer_size = 1M
low_priority_updates=1
concurrent_insert=2
symbolic-links=0
thread_cache_size=4
query_cache_size=256M
query_cache_type=1
query_cache_limit=4M
log-bin=/disk2/mysql/logs/mysql-bin
expire_logs_days=1
sync_binlog=0
server-id       = 1
innodb_buffer_pool_size = 10G
innodb_flush_log_at_trx_commit = 2
innodb_file_per_table=1
innodb_flush_method=O_DIRECT
table_open_cache=4096
table_definition_cache=4096
innodb_log_buffer_size = 32M
innodb_log_file_size = 2047M

[mysqldump]
quick
max_allowed_packet = 128M

[mysql]
no-auto-rehash
max_allowed_packet = 128M

[myisamchk]
key_buffer_size = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M

[mysqlhotcopy]
interactive-timeout
[root@otrs ~]# 

Code: Select all

rpm -qa|grep otrs
otrs-5.0.10-01.noarch

Code: Select all

[root@otrs ~]# sysbench --db-driver=mysql --test=oltp --oltp-table-size=1000000 --mysql-db=test --mysql-user= --mysql-password= --max-time=60 --oltp-read-only=off --max-requests=0 --num-threads=32 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 32

Doing OLTP test.
Running mixed OLTP test
Using Special distribution (12 iterations,  1 pct of values are returned in 75 pct cases)
Using "BEGIN" for starting transactions
Using auto_inc on the id column
Threads started!
Time limit exceeded, exiting...
(last message repeated 31 times)
Done.

OLTP test statistics:
    queries performed:
        read:                            1362074
        write:                           486364
        other:                           194551
        total:                           2042989
    transactions:                        97260  (1620.68 per sec.)
    deadlocks:                           31     (0.52 per sec.)
    read/write requests:                 1848438 (30801.30 per sec.)
    other operations:                    194551 (3241.88 per sec.)

Test execution summary:
    total time:                          60.0117s
    total number of events:              97260
    total time taken by event execution: 1919.3066
    per-request statistics:
         min:                                  3.15ms
         avg:                                 19.73ms
         max:                                160.51ms
         approx.  95 percentile:              45.95ms

Threads fairness:
    events (avg/stddev):           3039.3750/35.44
    execution time (avg/stddev):   59.9783/0.00

[root@otrs ~]# 

error in mysq.error.log repeat every time when we open dashboard.
we checked parameters about max_allowed_packet and change to 128M, but this don't help us.

[reneeb]

This is not an OTRS issue... Is the MySQL on the same server as the OTRS? Did you check the possible causes mentioned on https://dev.mysql.com/doc/refman/5.7/en ... rrors.html ?

[InventoR]

yes, I tried to increase this value from:
wait_timeout was 28800, new:288000
interactive timeout was 28800, new 288000

I increased all values with *timeout* on multiply *10, but nothing changed.

OTRS and mysql on one server.

I changed to using socket:

Code: Select all

$Self->{DatabaseDSN} = "DBI:mysql:database=$Self->{Database};host=$Self->{DatabaseHost};mysql_socket=/var/lib/mysql/mysql.sock;";

[InventoR]

In other problem, I have ticket with over 50messages, and it loading very long time. Sometime I get error message with timeout from my webserver.

[EXG133]

I have had tickets with 15.000 articles thanks to mailloops that loaded without issue (to be fair, they loaded slowly). It sounds like your database is quite unstable, 50 should be no problem.

Verify your max_allowed_packets setting and if that looks good check https://www.percona.com/blog/2016/05/16 ... et-errors/