Login Failed : Process engine persistence exception

Hello, we deployed Camunda using docker and when trying to login we receive the following message: “Login Failed : Process engine persistence exception” . Full Stacktrace below:

Error querying database. Cause: java.sql.SQLException: Out of memory (Needed 4294967110 bytes)

The error may exist in org/camunda/bpm/engine/impl/mapping/entity/Authorization.xml

The error may involve defaultParameterMap

The error occurred while setting parameters

SQL: SELECT CASE WHEN EXISTS ( SELECT ID_ FROM ACT_RU_AUTHORIZATION WHERE TYPE_ = 2 AND ( USER_ID_ IN (?, ‘*’) OR GROUP_ID_ IN ( ? , ? , ? , ? ) ) ) THEN 1 ELSE 0 END

SQL: SELECT CASE WHEN EXISTS ( SELECT ID_ FROM ACT_RU_AUTHORIZATION WHERE TYPE_ = 2 AND ( USER_ID_ IN (?, ‘*’) OR GROUP_ID_ IN ( ? , ? , ? , ? ) ) ) THEN 1 ELSE 0 END

We checked attributes from database like

sort_buffer_size
read_buffer_size
read_rnd_buffer_size
join_buffer_size
max_connections
innodb_buffer_pool_size

All of them have reasonable values.
We are using camunda/camunda-bpm-platform:tomcat-7.13.0-SNAPSHOT image

Hi @devBunny,

I don’t recommend you to use a snapshot version, please choose one of the already released minor or alpha versions:
https://hub.docker.com/r/camunda/camunda-bpm-platform/tags
Also, please share with us the command of docker image creation.

Best regards,
Yana

1 Like

Hello, thanks for the tip. We just changed to a non snapshot version. we are using the image in AWS ECS with the following env variables:
DB_DRIVER: com.mysql.jdbc.Driver
DB_URL: jdbc:mysql://host:3306/database?useUnicode=yes&characterEncoding=UTF-8&autoReconnect=true
DB_VALIDATE_ON_BORROW:true
DB_VALIDATION_QUERY: SELECT 1
Also exposing port 8080

Does the issue still occur?

Please verify if the isolation level of the db is set correctly according to https://docs.camunda.org/manual/7.12/user-guide/process-engine/database/#isolation-level-configuration

Thank you Yana, we could find the issue. sort_buffer_size attribute was to high and causing some issues. But all your all tips were valid, we could improve our database connection. About isolation level, we are using READ_COMMITTED