Often the database import in Plesk hangs at 100%. Users often notice errors even while importing a comparatively small database.

At Bobcares, our Support Engineers often get the request to solve Plesk usage problems as a part of our Server Management Services.

Today, let’s analyze the cause and see how our Support Team fix Database import issues for our customers.

What causes Database import to hang in Plesk

Recently we received a request from one of our customers that the database import task from Plesk was getting hang. This was even the case with considerably smaller databases. Mysql error log at “/var/log/mysql/error.log” was reporting the error as shown below:

While we cross-checked the Mysql CPU usage, we noticed that the usage has spiked to a great extend. A typical usage that we noticed resembled the one below:

PID     USER   PR    NI   VIRT         RES       SHR   S    %CPU    %MEM    TIME+     COMMAND
16164  mysql   20     0    2654656   748200   8044  S     201.0      18.3   2698:08    mysqldCopy Code

This error generally triggers due to non-optimized client databases.

 

How to fix Database import hang in Plesk

As the database import tasks normally hang at 100% due to non-optimized databases, we need to optimize the databases to fix this error permanently.

One solution here would be to allocate RAM to the MySQL server. For this, we need to connect to the Plesk server via SSH. Then, we need to open the MySQL configuration file 

my.cnfCopy Code

 or 

my.iniCopy Code

with any text editor. Locations of these files vary for different Linux distros.

For CentOS/RHEL:

/etc/my.cnfCopy Code

For Debian/Ubuntu:

/etc/mysql/my.cnfCopy Code

Then, we need to add the following directives under the

[mysqld]Copy Code

 section or increase the values if these directives are already defined:

innodb_buffer_pool_size=1024M
query_cache_size=64MCopy Code

Once the changes are made, save the changes and close the file. Also, note to restart the MySQL service with the appropriate command:

For CentOS/RHEL:
# systemctl restart mariadb
For Debian/Ubuntu:
# systemctl restart mysqldCopy Code

Now, we need to monitor CPU usage. If CPU consumption increases again, our Support Engineers consider reviewing the SQL code of a database that has slow queries.

During a high level of CPU usage, we can find queries that are currently running and taking a lot of time with the command below:

plesk db "SHOW FULL PROCESSLIST"Copy Code

This would help us to note down the queries that are getting stuck up. We also note to check the MySQL error log file

/var/log/mysqld.logCopy Code

 for errors.

Another possible reason could be low RAM and disk space available in the server. We can check RAM and free disk space with the command below:

# free -h
# df -hCopy Code

[Need any further assistance to fix Plesk server errors? – We’re available 24*7 ]

Conclusion

In short, the database import task in Plesk fails due to non-optimized client databases. Today, we saw some tips that our Support Engineers use to optimize the databases.