I have a deployment happening for my integration testing for flowable and once in a while it will get stuck with a change lock that occurs over and over again. How can I avoid this?
2019-08-27 21:12:48.132 INFO 8 — [ main] liquibase.executor.jvm.JdbcExecutor : SELECT LOCKED FROM flowable.ACT_DMN_DATABASECHANGELOGLOCK WHERE ID=1
2019-08-27 21:12:48.134 INFO 8 — [ main] l.lockservice.StandardLockService : Waiting for changelog lock…
It could have a few reasons: connections limit reached, liquibase process killed in the middle, etc. However, eventually it should give a timeout (but might be very long depending on the database config).
What database are you using? Can you check some connection/transaction timeout settings?
We’re using AWS RDS with the MySQL backend. I’ll check the specs but I doubt there’s enough traffic to cause problems.
Is the lock timeout a database setting or is it set by the app?
Sorry for the delay, this forum post slipped through.
It looks like Mysql support a ‘wait_timeout’ settings. It seems AWS does support this: https://aws.amazon.com/blogs/database/best-practices-for-configuring-parameters-for-amazon-rds-for-mysql-part-3-parameters-related-to-security-operational-manageability-and-connectivity-timeout/
(but the default seems to be 4 hours)
However, I’m not sure where exactly this setting can be set in AWS (can’t find if it can be added to the jdbc url, too)