Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Bigtop >> mail # dev >> Apache Pig 0.11.1 release candidate


Copy link to this message
-
Apache Pig 0.11.1 release candidate
Hi all,
I am a contributor to Apache Bigtop <http://bigtop.apache.org> and have a
question for you.
Bigtop is a TLP responsible for performing packaging and interoperability
testing of various projects in the Hadoop ecosystem, including Apache Pig.

We are planning to include Pig 0.11 in our soon to be released Bigtop 0.6
distribution. However, while upgrading Pig from 0.10 to 0.11, I wasn't able
to compile Pig 0.11.1 on RPM based
systems<http://bigtop01.cloudera.org:8080/view/Bigtop-trunk/job/Bigtop-trunk-Pig/313/label=centos6/console>.
There doesn't seem to be anything Bigtop specific here, I would expect this
issue to impact all Pig users. It seems like Pig's contrib sub-project uses
the system's default encoding for compiling code; however on RPM based
systems, the default encoding is not suitable and breaks the build. I
created PIG-3262 <https://issues.apache.org/jira/browse/PIG-3262> to track
this and Cheolsoo graciously committed this to Pig trunk. The essence of
Bigtop is exactly to find integration issues like this.

Now, I do realize that Bill and the community has done some excellent work
in putting together 0.11.1. Perhaps, I am a little too late to ask this
question but I thought I'd ask it anyway. Is there a possibility that the
Pig community can release a new release candidate for 0.11.1 with the fix
in PIG-3262?

The pros:
1. It would allow Pig users to compile Pig contrib on RPM machines
(RHEL/CentOS 5, 6, SLES 11, Fedora, etc.) which doesn't seem to be possible
as of now.
2. It would enable Apache Bigtop 0.6 to include a Pig version that builds
on all OS variants.

The cons:
1. There is a cost of cutting out another release candidate to the Pig
community. I completely understand and appreciate the cost involved;
however, I would anticipate the cost to be minimal since a) the
change<https://issues.apache.org/jira/secure/attachment/12575962/PIG-3262.2.patch>is
quite trivial; b) the change only affects the contrib functionality
and
not the "core" functionality, per se.

If we do decide to release another release candidate, I would be more than
happy to perform integration testing on it by means of Apache Bigtop.

I do realize the unfortunate timing of this email, it would have been ideal
if we were having this conversation a week ago while the vote was still
going on. I will try to change that in future so please do accept my
apologies in advance.

Regards,
Mark
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB