Yes, I think we should have randomness (or some rotation) in picking Hadoop-1 vs Hadoop-2. This will not eliminate the problem, but contain it at a tolerable minimum.
Otherwise, this sort of build failure will keep coming. Thanks, Xuefu On Thu, Sep 25, 2014 at 7:17 PM, Navis류승우 <navis....@nexr.com> wrote: > Hi, > > I've booked current build failure problem on HIVE-8265. It seemed this kind > of things will happen again. > > Thanks, > Navis > > > 2014-09-26 10:40 GMT+09:00 Szehon Ho <sze...@cloudera.com>: > > > Hi all, > > > > There's been no build coverage of hadoop-1 on hive-trunk since 0.13 > > release. I'm planning to restart periodic hadoop-1 builds at > > > > > http://ec2-174-129-184-35.compute-1.amazonaws.com/jenkins/job/HIVE-TRUNK-HADOOP-1 > > , > > so we know the current state. I hope this helps for the 0.14 release and > > beyond. > > > > But as I'm setting it up, there's an issue compiling with "mvn install > > -Phadoop-1" on the hive trunk root dir, I wonder if this is still the > > command to compile it, and it is a real error? > > > > [ERROR] Failed to execute goal > > org.apache.maven.plugins:maven-compiler-plugin:2.3.2:compile > > (default-compile) on project hive-exec: Compilation failure > > [ERROR] > > > /data/hive-ptest/working/apache-svn-trunk-source/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/CustomPartitionVertex.java:[37,27] > > error: cannot find symbol > > > > > > Appreciate any help, thanks. > > Szehon > > >
