I've been dealing with a similar situation and I haven't found other solution rather than launching two independent jobs (with a script or whatever you like), letting the output of the first be the input of the last. If you find any other option please let me know.
Regards On 12 February 2014 12:55, Massimo Simoniello <[email protected]>wrote: > Hi, > > I'm using Hadoop Pipes and I want to chain two jobs (job1 and job2). Is it > possible? > I use the FileInputFormat.addInputPath() > and FileOutputFormat.setOutputPath() functions to do it in Java, but I want > to know if there is some way for do it in C++ with pipes. > > Thanks in advance, > > Massimo > > > >
