How are you starting Flume? What platform / environment are you running on? Did you write your own init scripts or are you using a vendor Hadoop distribution (i.e. Cloudera) or something else (i.e. directly using Bigtop)? On Linux, if you are writing your own init scripts then running Flume via nohup might help with the signal handling.
Mike Sent from my iPhone > On Mar 23, 2014, at 11:15 PM, lulynn_2008 <lulynn_2...@163.com> wrote: > > Thanks. Do you know how to find the things which throw "a SIGHUP" or "a > SIGTERM"? > At the beginning, flume worked normally. After we made some actions, flume > becomes unreliable. There must be something which made flume shut down, > right? Is it possible to do something to make flume work again? Like stop > some kinds of services or kill some kinds of sessions? > > > > At 2014-03-21 19:46:13,"Christopher Shannon" <cshannon...@gmail.com> wrote: > I have also experienced this. A SIGHUP or a SIGTERM will gracefully shut it > down. So look for anything in your system throwing those. Pretty much any > other signal will kill it outright. > >> On Friday, March 21, 2014, lulynn_2008 <lulynn_2...@163.com> wrote: >> Hi All, >> >> After flume agent is started at 1:10 and it shut itself down at 2:08. No >> errors, but graceful shutdown. This situation has happened several times. >> My question is what will possibly gracefully shut down flume? Or which side >> of environment should I pay attention to to trace the error or find the root >> cause? >> >> Thanks > >
