RHEA-2014:1275 - Product Enhancement Advisory

Topic

Updated rhs-hadoop and rhs-hadoop-install packages that adds many enhancements and fix multiple bugs are now available for Red Hat Storage 3.

Description

Red Hat Storage is a software only scale-out storage solution that provides
flexible and affordable unstructured data storage. Red Hat Storage 3.0
unifies data storage and infrastructure, increases performance, and
improves availability and manageability to meet enterprise-level storage
challenges. Red Hat Storage provides compatibility for Apache Hadoop and
it uses the standard file system APIs available in Hadoop to provide a
new storage option for Hadoop deployments.

• rhs-hadoop component is the Hadoop file system plugin for Red Hat Storage

3.0. It has undergone a number of recent changes in order to add support
for the use of more than one RHS volume with Hadoop.

• rhs-hadoop-install component is the installer for rhs-hadoop for Red Hat

Storage 3.0. It has undergone a significant re-write to support both new
and existing Red Hat Storage clusters as well as support the use of
multiple volumes.

This advisory adds many enhancements and fix multiple bugs:

• Previously, executing script setup_cluster.sh script copied all files to

/tmp/bin directory instead of using the existing scripts located in the
node from rhs-hadoop-install package. Now, with this release, the files
are not copied and you must run "yum install rhs-hadoop-install" command
on each node.(BZ#1131171)

• Previously, the rhs-high-thoughput profile was not enabled by the

setup_cluster.sh script. Now, with this release, --profile allows enabling
rhs-high-thoughput profile by setup_cluster.sh script. (BZ#1139458)

• The TEZ service is now supported from Hortonworks Data Platform 2.1 in

Red Hat Storage 3.0 for Hadoop stack. (BZ#1139710)

• The HBase service is now supported on Hortonworks Data Platform 2.1 in

Red Hat Storage 3.0 for Hadoop workloads. (BZ#1139717)

• Previously, the default volume for Hadoop was not defined in

enable_vol.sh. Now, with this release, specifying "--make-default"
in enable_vol.sh enables the specified volume to be the default volume
for Hadoop. (BZ#1141348)

• Previously, executing setup_cluster.sh script reinstalled Ambari even

if Ambari was running. Now, with this release, executing setup_cluster.sh
script does not reinstall Ambari. (BZ#1151211)

• A new option has been introduced that enables users to specify an Ambari

repository URL which can be different from the repository URL used by
setup_cluster.sh. With this fix, --ambari-repo <url> option allows users
to install and use new Ambari repository files. (BZ#1151215)

• A new auxillary script has been added to bin directory which supports

all actions to modify hadoop configuration files. With this script,
users can now add or delete core-site keys and modify core-site property
values. (BZ#1151219)

All users of Red Hat Storage are advised to install these updated packages.

Solution

Before applying this update, make sure all previously released errata
relevant to your system have been applied.

This update is available via the Red Hat Network. Details on how to
use the Red Hat Network to apply this update are available at
https://access.redhat.com/articles/11258

Affected Products

• Red Hat Gluster Storage Server for On-premise 3 for RHEL 6 x86_64

Fixes

• BZ - RHEA-2014:1275 - missing apps/hive/warehouse
• BZ - RHEA-2014:1275 - [BREW] rhs-hadoop-2.1.6-2 rpmlint check - some minor errors
• BZ - RHEA-2014:1275 - rhs-hadoop-install deletes files in the gluster volume and the volume itself.
• BZ - RHEA-2014:1275 - UIDs across cluster not verified on mgmt-node when it's outside the storage pool
• BZ - RHEA-2014:1275 - install.sh should never delete files in the gluster volume
• BZ - RHEA-2014:1275 - install.sh does not handle the case of "hadoop" group users in parsing "getent group hadoop"
• BZ - RHEA-2014:1275 - add hive and hcat users; add app/ and app/webhcat/ directories
• BZ - RHEA-2014:1275 - NullPointerException is mapreduce.jobtracker.system.dir undefined.
• BZ - RHEA-2014:1275 - wrong error message for mount points configuration
• BZ - RHEA-2014:1275 - listStatus corrupts : file paths with % and other URI Escape encodings are corrupted
• BZ - RHEA-2014:1275 - Gluster file names with special characters not seen by hadoop fs -ls
• BZ - RHEA-2014:1275 - required users created by setup_cluster.sh do not have same uid across cluster
• BZ - RHEA-2014:1275 - create_vol.sh should do all validation(wrong argument count etc) before running volume create command
• BZ - RHEA-2014:1275 - create_vol.sh do volume mount multiple times if it has more than one brick.
• BZ - RHEA-2014:1275 - rhs-node listed as rhs_node in enable_vol.sh help text
• BZ - RHEA-2014:1275 - [doc] Need a Configuration/Installation steps/script in case of removing brick(and also removing node from cluster) from hadoop enabled volume and cluster.
• BZ - RHEA-2014:1275 - some problems in rpmlint check output
• BZ - RHEA-2014:1275 - [BREW] incorrect package content description
• BZ - RHEA-2014:1275 - add to the installer rhs-high-throughput profile, on each storage node
• BZ - RHEA-2014:1275 - [RFE] Support HBase from HDP 2.1 in RHSS 3.0+
• BZ - RHEA-2014:1275 - [RFE] add ability to define the default volume in enable_vol.sh
• BZ - RHEA-2014:1275 - [RFE] setup_cluster should not, by default, yum install ambari (agent/server) when already installed
• BZ - RHEA-2014:1275 - [RFE] complete the actions supported by bin/ambari_config_update.sh (add, delete, replace, remove)

CVEs

(none)

References

• http://www.redhat.com/security/updates/classification/#normal