Claus Ibsen created CAMEL-25754:
-----------------------------------
Summary: camel-test-infra-artemis - no disk usage limit, so a
broker on a nearly full Docker disk does not block producers
Key: CAMEL-25754
URL: https://issues.apache.org/jira/browse/CAMEL-25754
Project: Camel
Issue Type: Improvement
Components: camel-test-infra
Reporter: Claus Ibsen
Fix For: 4.24.0
camel infra run artemis starts the Artemis container with the broker's default
max-disk-usage of 90%. On a developer machine, the disk that counts is the
Docker VM disk, which often fills with images. Above 90%, the broker blocks
every producer:
{noformat}
WARN [org.apache.activemq.artemis.core.server] AMQ222210: Free storage space
is at 2.3GB of 84.4GB total. Usage rate is 97.3% which is beyond the configured
<max-disk-usage>. System will start blocking producers.
WARN [org.apache.activemq.artemis.core.server] AMQ222212: Disk Full! Blocking
message production on address 'orders'. Clients will report blocked.
{noformat}
The Camel route then hangs on its send. Its own log shows only a WARN from the
Artemis client ("AMQ212054: Destination address=orders is blocked"), and the
messages never arrive. For development and tests, a broker should not need many
GB of free disk space.
The embedded Artemis services in camel-test-infra-artemis already raise the
limit (setMaxDiskUsage(98)); the container (ArtemisContainer) sets nothing.
Fix: switch the disk check off (-1) in the container
(JAVA_ARGS_APPEND=-Dbrokerconfig.maxDiskUsage=-1, checked with the same image:
no more blocking at 98% disk usage) and in the embedded services.
Found while building a JMS example for the AI benchmark.
_Claude Code on behalf of davsclaus_
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
