Yes that's what I wanted to do, but that feedback loop is killer. I'll just have to switch to something else.
Thanks - Connor On Sat, Jan 19, 2013 at 5:25 PM, Brock Noland <[email protected]> wrote: > Hi, > > Do I understand this correctly, you are going to use the flume log4j > appender to collect flume logs? > > If so, I don't see how you'd avoid the feedback loop. > > Brock > > On Fri, Jan 18, 2013 at 11:13 AM, Connor Woodson <[email protected]> > wrote: > > I just ran into an unfortunate configuration issue; I want to use flume's > > log4j appender as part of the agent itself in order to send error logs > to a > > monitoring system. However, when that system is down (which it should be > > able to be without causing issues), the RPCClient (even though it's in a > > Failover Sink) throws an error/warning that it can't connect to the > system, > > and then that error gets routed recursively back into the appender and so > > forth... > > > > Is there any way to make this system work (using the log4j appender on > error > > messages), or if not any recommendations on how to track my error logs? > > > > Thanks, > > > > - Connor > > > > -- > Apache MRUnit - Unit testing MapReduce - > http://incubator.apache.org/mrunit/ >
