[ 
https://issues.apache.org/jira/browse/LOG4J2-2895?focusedWorklogId=459515&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-459515
 ]

ASF GitHub Bot logged work on LOG4J2-2895:
------------------------------------------

                Author: ASF GitHub Bot
            Created on: 15/Jul/20 20:25
            Start Date: 15/Jul/20 20:25
    Worklog Time Spent: 10m 
      Work Description: carterkozak opened a new pull request #389:
URL: https://github.com/apache/logging-log4j2/pull/389


   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]


Issue Time Tracking
-------------------

            Worklog Id:     (was: 459515)
    Remaining Estimate: 0h
            Time Spent: 10m

> Async logging should avoid submitting work from Log4jThread threads
> -------------------------------------------------------------------
>
>                 Key: LOG4J2-2895
>                 URL: https://issues.apache.org/jira/browse/LOG4J2-2895
>             Project: Log4j 2
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 2.13.3
>            Reporter: Carter Kozak
>            Assignee: Carter Kozak
>            Priority: Major
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> In order to avoid deadlocks, when we have the option we should try to avoid 
> enqueueing events for async loggers from threads that extend Log4jThread. 
> Submitting events between AsyncLoggerContexts is rare, but can allow multiple 
> disruptors to deadlock feeding events from one to the other. Similarly using 
> both mixed async loggers with AsyncLoggerContext results in events flowing 
> from one disruptor to another.
> There are several areas we can avoid these potential lockups by checking if 
> the current thread is a Log4jThread and logging synchronously. Note that in 
> these cases, events may be serialized to disk in an unexpected order, however 
> the scenario is both sufficiently rare and dangerous that I think it's 
> worthwhile to change.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to