GitHub user tcrayford opened a pull request:
https://github.com/apache/kafka/pull/1614
KAFKA-3933: close deepIterator during log recovery
Avoids leaking native memory and hence crashing brokers on bootup due to
running out of memory.
Introduces `kafka.common.ClosableIterator`, which is an iterator that
can be closed, and changes the signature of
`ByteBufferMessageSet.deepIterator` to return it, then changes the
callers to always close the iterator.
This is a followup from https://github.com/apache/kafka/pull/1598 with more
native memory leaks in the broker code found and fixed.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/heroku/kafka dont_leak_native_memory
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/kafka/pull/1614.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #1614
----
commit 7fa608371ff8e380350029f3ab1dcffeb0e26c73
Author: Tom Crayford <[email protected]>
Date: 2016-07-08T11:50:21Z
KAFKA-3933: close deepIterator during log recovery
Avoids leaking native memory and hence crashing brokers on bootup due to
running out of memory.
Introduces `kafka.common.ClosableIterator`, which is an iterator that
can be closed, and changes the signature of
`ByteBufferMessageSet.deepIterator` to return it, then changes the
callers to always close the iterator.
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
