Sophie Blee-Goldman created KAFKA-10262:
-------------------------------------------
Summary: StateDirectory is not thread-safe
Key: KAFKA-10262
URL: https://issues.apache.org/jira/browse/KAFKA-10262
Project: Kafka
Issue Type: Bug
Components: streams
Affects Versions: 2.6.0
Reporter: Sophie Blee-Goldman
As explicitly stated in the StateDirectory javadocs, "This class is not
thread-safe."
Despite this, a single StateDirectory is shared among all the StreamThreads of
a client. Some of the more "dangerous" methods are indeed synchronized, but
others are not. For example, the innocent-sounding #directoryForTask is not
thread-safe and is called in a number of places. We call it during task
creation, and we call it during task closure when we check
`directoryForTaskIsEmpty`. It's not uncommon for one thread to be closing a
task while another is creating it after a rebalance.
In fact, we saw exactly that happen in our test application. This ultimately
lead to the following exception
{code:java}
org.apache.kafka.streams.errors.ProcessorStateException: task directory
[/mnt/run/streams/state/stream-soak-test/1_0] doesn't exist and couldn't be
created at
org.apache.kafka.streams.processor.internals.StateDirectory.directoryForTask(StateDirectory.java:112)
at
org.apache.kafka.streams.processor.internals.ProcessorStateManager.<init>(ProcessorStateManager.java:187)
at
org.apache.kafka.streams.processor.internals.StandbyTaskCreator.createTasks(StandbyTaskCreator.java:85)
at
org.apache.kafka.streams.processor.internals.TaskManager.handleAssignment(TaskManager.java:337)
{code}
The exception arises from this line in StateDirectory#directoryForTask:
{code:java}
if (hasPersistentStores && !taskDir.exists() && !taskDir.mkdir())
{code}
Presumably, if the taskDir did not exist when the two threads began this
method, then they would both attempt to create the directory. One of them will
get there first, leaving the other to return unsuccessfully from mkdir and
ultimately throw the above ProcessorStateException.
I've only confirmed that this affects 2.6 so far, but the unsafe methods are
present in earlier versions. It's possible we made the problem worse somehow
during "The Refactor" so that it's easier to hit this race condition.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
