I'm redoing everything on my end.... pretty much from scratch.... The question I had when I woke up this morning... Do I really need to configure VM's in order to run multiple nodes? Or does running multiple, Nutch-Solr nodes, sufice if they are under their own user space? If I remember correct;ly one could not run multiple Nutch instances from the same user because all those instances would write to the same TMP file or something like this... But this shouldn't be the case when each instance is run from its own user account.
With that sayd... If its really as easy as running the instances from their own account... I could try and use Hadoop 3, say like till the end of next week and report back. Greetz On Wed, Mar 14, 2018 at 12:01 AM, Lewis John McGibbney <lewi...@apache.org> wrote: > Hi Seb, > > On 2018/03/12 11:00:52, Sebastian Nagel <wastl.na...@googlemail.com> > wrote: > > Hi, > > > > > seeing as we have just merged in the 'new' MR patch > > > > yep, but there's still something to do (NUTCH-2517, > > ACK, this needs more testing. > > > NUTCH-2518). > > I honestly didn't see this come through but yes you are right. > > > Better to address this before any upgrade of the Hadoop version. > > ACK > > > But since there seem to be no breaking MapReduce API changes > > http://hadoop.apache.org/docs/r3.0.0/index.html > > I would even expect that the Nutch job jar (built for 2.7) > > will run on Hadoop 3.0, or does it not? > > > > I have absolutely no idea. I've certainly not had an opportunity to run on > H v3 cluster. >
