[
https://issues.apache.org/jira/browse/LOG4J2-1397?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15322161#comment-15322161
]
Anthony Maire commented on LOG4J2-1397:
---------------------------------------
It just doesn't work because of different orderings in the preallocated array
and the ringbuffer. Let's take a simple example:
- 2 threads
- a ringbuffer with 2 slots
So we need a preallocated array with 4 buffers
- Thread A get sequence #1 from the atomic counter, then get interrupted by the
system scheduler before doing the copy
- Thread B is still logging, it get sequence #2, #3, #4 from the counter,
pushed 2 events to the async logger (so the ringbuffer is full) and is blocked
trying to push the third
- Async Logger processed the first event submitted by thread B
- Thread B can push another event (with preallocated buffer sequence #4)
- Thread B wants to log another event, it get sequence #5 for the preallocated
array
- Thread A is resumed by the scheduler
=> Data race, both thread will copy data in the same preAllocated buffer
> Support ByteBufferLayout
> ------------------------
>
> Key: LOG4J2-1397
> URL: https://issues.apache.org/jira/browse/LOG4J2-1397
> Project: Log4j 2
> Issue Type: New Feature
> Components: Layouts
> Affects Versions: 2.6
> Reporter: Remko Popma
>
> As requested by Kirk Pepperdine on the Mechanical Sympathy [mailing
> list|https://groups.google.com/d/msg/mechanical-sympathy/klefjRqlpQE/-1WF59IGAwAJ]:
> Support a new Layout that takes ObjectMessages containing a ByteBuffer
> payload and write them to the appender without modification.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
