Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [geomesa-users] plan for supporting newer accumulo releases?

Thanks Jim!

I am very interested in leveraging the latest kerberos auth support of accumulo 1.7:
https://issues.apache.org/jira/browse/ACCUMULO-2815
Right now, we have to pass in accumulo user/pass to run a geomesa program, which is not very desirable.

I've tried accumulo 1.6 and that didn't work with geomesa master branch, clearly due to dependency issues. I haven't tried that branch yet, but directly to 1.7 is certainly better :-)

I am running CDH stack for hadoop/yarn, currently on 5.3.x release, which comes with zookeeper 3.4.5 and hadoop 2.5.0. CDH 5.4.x is also available, and it's based on hadoop 2.6.

In general, I like how spark handles dependency - you can package almost any version of hadoop client. For example, I built my spark setup with commands like:
mvn -Pyarn -Phadoop-2.4 -Dhadoop.version=2.5.0-cdh5.3.3 -DskipTests clean package
http://spark.apache.org/docs/latest/building-spark.html#specifying-the-hadoop-version

I'd be interested to know how you guys decide to go next and potentially help with some testing. But so far, our geomesa setup looks promising. Great job!

-Simon




On Fri, May 29, 2015 at 3:04 PM, Jim Hughes <jnh5y@xxxxxxxx> wrote:
Hi Simon,

Great question.  At the moment, we've been focused on Accumulo 1.5, and provided some support for 1.6 as the need has come up. 

There are some branches which support Accumulo 1.6 up on GitHub.  Of note, a1.6_h.2.6 handles some of the changes which are needed for Accumulo 1.6 and Hadoop 2.5/2.6. 

GeoMesa is at the exciting corner of several dependencies, so managing a complete set of everything one is a tad tricky.  I do want to provide better support as we continue.  In general, I'm happy to build a branch for a requested set of versions (Accumulo, Hadoop, Zookeeper).

Do you have a present need for GeoMesa to be capability with Accumulo 1.6 or 1.7?  If so, could you suggest some versions of Hadoop and Zookeeper which may be associated?  If so, I'm happy to build up a branch, and shake down any api/unit tests changes.  In exchange, I'd ask for some help identifying any gotchas while deploying iterators or the GeoServer plugin.

Cheers,

Jim


On 05/29/2015 02:44 PM, Xu (Simon) Chen wrote:
Hey folks,

I believe only accumulo 1.5 works with geomesa, as I previously had no luck with accumulo 1.6. And now accumulo 1.7 is released, which supports client kerberos auth - that's a valuable enterprise feature...

I am wondering if there is a plan to move geomesa to newer accumulo releases?

Thanks.
-Simon


_______________________________________________
geomesa-users mailing list
geomesa-users@xxxxxxxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
http://www.locationtech.org/mailman/listinfo/geomesa-users


_______________________________________________
geomesa-users mailing list
geomesa-users@xxxxxxxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
http://www.locationtech.org/mailman/listinfo/geomesa-users


Back to the top