You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We use testcontainers in spring-based microservices mostly for launching PostgresQL database. It is enough to use jdbc-url based configuration for us. Also usually we run containers in daemon mode.
Sometimes when I run tests on my machine and then interrupt them (with Ctrl+C, for example), launched container doesn't stop. This is expected because process didn't stop gracefully.
But while it is ok on local machine, it becomes a trouble on CI agents. I noticed many zombie postgres containers on CI agent machines. Probably it hapens because of OutOfMemory on agents (it happens seldom, but it does), probably because of #253 (see the last comment) or something else.
Anyway it seems useful to store launched containers to have an ability to stop them automatically in exceptional cases.
The text was updated successfully, but these errors were encountered:
Did you try version 1.6.0? There is a new feature called "Death note" ( See #545 ) for that. Even if you kill -9 your process, the containers will get cleaned.
FYI there is also org.testcontainers=true label on containers started with Testcontainers
Hi, thank you for awesom development tool.
We use testcontainers in spring-based microservices mostly for launching PostgresQL database. It is enough to use jdbc-url based configuration for us. Also usually we run containers in daemon mode.
Sometimes when I run tests on my machine and then interrupt them (with Ctrl+C, for example), launched container doesn't stop. This is expected because process didn't stop gracefully.
But while it is ok on local machine, it becomes a trouble on CI agents. I noticed many zombie postgres containers on CI agent machines. Probably it hapens because of OutOfMemory on agents (it happens seldom, but it does), probably because of #253 (see the last comment) or something else.
Anyway it seems useful to store launched containers to have an ability to stop them automatically in exceptional cases.
The text was updated successfully, but these errors were encountered: