Yes, the tests write in /tmp & in a subdirectory of the builds. Having a different user for dev & prod would secure this as well ;-)
On Thu, Nov 8, 2012 at 1:04 PM, Jean-Marc Spaggiari <jean-m...@spaggiari.org > wrote: > Thanks for the reply. > > That was all what was thinking about (port number, directory, bugs, > etc.) regarding running the tetsts on the local node... So seems it's > a good thing I stoped them. > > If I stop my local datanode/region server, then bring down my network > interface before I run the tests, I should be fine, right? I'm running > the tests with my own user while hbase and hadoop are both running for > a their own. I don't think there is a risk for the local test to fine > and write on the local HBase/Hadoop files. > > JM > > 2012/11/8, Nicolas Liochon <nkey...@gmail.com>: > > Today, at this time of writing, it should be ok: the tests write in a > > specific directory, and the ports are dynamic. But it's not without > danger: > > what if tomorrow someone creates a bug and hardcodes a port already used > by > > your cluster ? It's unlikely; but it happened in the past. As well, the > CPU > > & memory used by the tests won't be available to your datanode / > > regionserver, so you doing a kind of load test on this node... > > > > N. > > > > > > On Thu, Nov 8, 2012 at 12:26 PM, Jean-Marc Spaggiari < > > jean-m...@spaggiari.org> wrote: > > > >> Hi, > >> > >> My computer is hosting a region server and a datanode for a 8 nodes > >> cluster. > >> > >> Now, I would like to run the JUnit tests locally too. > >> > >> Is there going to be any conflict between my local regionserver and > >> the HBase tests? Is there any risk for my data if I do that? > >> > >> Even if I turn my local region server off, since there is already a > >> zookeeper on the network, will again the local test be a risk for the > >> cluster? > >> > >> Thanks, > >> > >> JM > >> > > >
