Thanks. I've known for months that the changes Flume was making for the embedded support are not backward compatible. So when we upgrade, the embedded appender will require a minimum of Flume 1.4. However, the avro and persistent variations should still work with 1.2, 1.3 and 1.3.1.
I don't know if this will make it into the next beta. At the moment I am working on the nasty issue with Berkely DB that Edward reported. Luckily, very few applications are likely to be affected by the problem. Ralph On Jul 2, 2013, at 12:15 PM, Edward Sargisson wrote: > Hi Gary, > The easiest reference I know of is Jira: > https://issues.apache.org/jira/browse/FLUME-1630 > Log4j2's embedded agent implementation uses APIs that aren't really publicly > documented. That code did get adopted by Flume and is available as embedded > agent in Flume JARS. > http://flume.apache.org/releases/content/1.4.0/FlumeDeveloperGuide.html#embedded-agent > > Using the Flume embedded agent for the Log4j2 agent is possibly the best > approach: it's the original log4j2 code slightly modified and should continue > to be maintained by the Flume team. Hopefully there are not problems where it > doesn't actually work. > > Cheers, > Edward > > > On Tue, Jul 2, 2013 at 11:29 AM, Gary Gregory <[email protected]> wrote: > On Tue, Jul 2, 2013 at 2:17 PM, Edward Sargisson <[email protected]> wrote: > Hi all, > You may be interested to know that Flume 1.4.0 was released this morning. > > This is relevant to Log4j2 to because the Log4j2 Embedded Flume Appender > relies on classes in Flume 1.3.1 for configuration that have changed in Flume > 1.4.0. > > > Thank you for the heads up. Where are these changed doc'd? > > Gary > > Cheers, > Edward > > > > > -- > E-Mail: [email protected] | [email protected] > Java Persistence with Hibernate, Second Edition > JUnit in Action, Second Edition > Spring Batch in Action > Blog: http://garygregory.wordpress.com > Home: http://garygregory.com/ > Tweet! http://twitter.com/GaryGregory >
