While analyzing a connection problem between a client and a MariaDB 10.2 server, the so-called "error log" needed to be checked for warnings and errors. However the file (/var/log/mysql/error.log) did not contain any logged events.
First thing to verify is the MySQL/MariaDB config whether or not the error_log was actually enabled. On this machine, running MariaDB 10.2 on Ubuntu 18.04 (Bionic), the config files seemed in order:
ck@mariadb:~$ cat /etc/mysql/conf.d/mariadb.cnf | grep error
# log errors
But the running config needs to be verified as well. This can be done by checking the global variables:
Welcome to the MariaDB monitor. Commands end with ; or \g.
Your MariaDB connection id is 308342315
Server version: 10.2.28-MariaDB-1:10.2.28+maria~bionic-log mariadb.org binary distribution
Copyright (c) 2000, 2018, Oracle, MariaDB Corporation Ab and others.
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
MariaDB [(none)]> show global variables where variable_name like '%error%';
| Variable_name | Value |
| log_error | /var/log/mysql/error.log |
| max_connect_errors | 100 |
| max_error_count | 64 |
| slave_skip_errors | OFF |
4 rows in set (0.00 sec)
Same here: The log_error variable is set to a path - which means error logging is enabled.
Another potential reason of failure could be that the file does not exist, or that the permissions are wrong. The MySQL/MariaDB process, running usually under the user "mysql", needs to be able to write into this file:
root@mariadb:/var/log/mysql# ls -la
-rw-r----- 1 mysql adm 0 Oct 20 00:08 error.log
-rw-r----- 1 mysql adm 0 Oct 19 00:07 error.log.1
-rw-r----- 1 mysql adm 5037450 Apr 10 2020 error.log.8.gz
-rw-rw---- 1 mysql adm 105161366 Oct 18 11:58 mariadb-bin.007219
-rw-rw---- 1 mysql adm 105400817 Oct 18 13:26 mariadb-bin.007220
Inside the directory /var/log/mysql/ the files are listed. The error.log file clearly exists, and with the correct permissions; the mysql user is owner and can therefore read and write this file. However the size is 0 - meaning the file is empty and nothing was logged.
The fact that the MariaDB process is able to successfully write the binary logs (which are enabled on this machine, too) in the same path, shows that there is definitely no permission problem.
It turns out that with the MariaDB package, an additional config file was installed, redirecting the error log to syslog under the [mysqld_safe] config section:
admck@mariadb:~$ cat /etc/mysql/conf.d/mysqld_safe_syslog.cnf
This config file was installed by the mariadb-server package:
admck@mariadb:~$ dpkg -S /etc/mysql/conf.d/mysqld_safe_syslog.cnf
As Ubuntu starts MySQL/MariaDB with mysqld_safe (which can be seen in the process list), these config options are read and applied during the start of MariaDB. They cause the following behaviour:
It's important to note that log_error and syslog cannot co-exist! Otherwise a warning will be issued and log_error takes precedence.
Once these two config options were commented/disabled and MariaDB restarted, the /var/log/mysql/error.log now started to log warnings and errors:
admck@mariadb:~$ ls -la /var/log/mysql/error.log
-rw-r----- 1 mysql adm 953061 Oct 20 17:14 /var/log/mysql/error.log
No comments yet.
AWS Android Ansible Apache Apple Atlassian BSD Backup Bash Bluecoat CMS Chef Cloud Coding Consul Container Containers CouchDB DB DNS Database Databases Docker ELK Elasticsearch Filebeat FreeBSD Galera GlusterFS Grafana Graphics HAProxy HTML Hacks Hardware Icinga Icingaweb Icingaweb2 InfluxDB Internet Java KVM Kibana Kodi Kubernetes LXC Linux Logstash Mac Macintosh Mail MariaDB Minio MongoDB Monitoring Multimedia MySQL NFS Nagios Network Nginx OSSEC OTRS Office PGSQL PHP Perl Personal PostgreSQL Postgres PowerDNS Proxmox Proxy Python Rancher Rant Redis Roundcube SSL Samba Seafile Security Shell SmartOS Solaris Surveillance Systemd TLS Tomcat Ubuntu Unix VMWare VMware Varnish Virtualization Windows Wireless Wordpress Wyse ZFS Zoneminder