hadoop-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ajay Kumar <ajay.ku...@cloudera.com.INVALID>
Subject FW: VOTE: Hadoop Ozone 0.4.0-alpha RC2
Date Tue, 07 May 2019 16:13:50 GMT
Thanks for trying out ozone-0.4.0 RC2 release artifacts and for your votes.

The vote is PASSED with the following details:
*  3 binding +1, (thanks 	 Arpit, Xiaoyu & Jitendra)
*  4 non-binding +1, (thanks Eric, Dinesh, Marton) together with my closing +1 [*]
*  no -1/0

Thanks again, will publish the release now.

Note: ** We have addressed the issue highlighted by Xiaoyu, renaming "hadoop-3.3.0-SNAPSHOT-src-with-hdds"
to " hadoop-ozone-0.4.0-alpha-src". 
Git hash still remains same. MD5 checksum has been updated for renamed file. Release documentation
for ozone has been updated to handle this in future releases.


On 5/5/19, 8:05 PM, "Xiaoyu Yao" <xyao@cloudera.com> wrote:

    +1 Binding. Thanks all who contributed to the release. 
    + Download sources and verify signature.
    + Build from source and ran docker-based ad-hot security tests.
    ++ From 1 datanode scale to 3 datanodes, verify certificates were correctly issued when
security enabled
    ++ Smoke test for both non-secure and secure mode.
    ++ Put/Get/Delete/Rename Key with 
    +++ Kerberos testing
    +++ Delegation token testing with DTUtil CLI and MR jobs.
    +++ S3 token.
    Just have one minor question for the expanded source code which points to hadoop-3.3.0-SNAPSHOT-src-with-hdds/hadoop-ozone.
But in hadoop-ozone/pom.xml, we explicitly declare dependency on Hadoop 3.2.0.
    I understand we just take the trunk source code(3.3.0-SNAPSHOT up to the ozone-0.4 RC)
here, should we fix this by giving the git hash of the trunk or clarify it to avoid confusion?

    This might be done by just updating the name of the binaries without reset the release
    On 5/3/19, 4:07 PM, "Dinesh Chitlangia" <dchitlangia@hortonworks.com> wrote:
        +1 (non-binding)
        - Built from sources and ran smoke test
        - Verified all checksums
        - Toggled audit log and verified audit parser tool
        Thanks Ajay for organizing the release.
        On 5/3/19, 5:42 PM, "Eric Yang" <eyang@hortonworks.com> wrote:
            On 4/29/19, 9:05 PM, "Ajay Kumar" <ajay.kumar@hortonworks.com> wrote:
                Hi All,
                We have created the third release candidate (RC2) for Apache Hadoop Ozone
                This release contains security payload for Ozone. Below are some important
features in it:
                  *   Hadoop Delegation Tokens and Block Tokens supported for Ozone.
                  *   Transparent Data Encryption (TDE) Support - Allows data blocks to be
                  *   Kerberos support for Ozone.
                  *   Certificate Infrastructure for Ozone  - Tokens use PKI instead of shared
                  *   Datanode to Datanode communication secured via mutual TLS.
                  *   Ability secure ozone cluster that works with Yarn, Hive, and Spark.
                  *   Skaffold support to deploy Ozone clusters on K8s.
                  *   Support S3 Authentication Mechanisms like - S3 v4 Authentication protocol.
                  *   S3 Gateway supports Multipart upload.
                  *   S3A file system is tested and supported.
                  *   Support for Tracing and Profiling for all Ozone components.
                  *   Audit Support - including Audit Parser tools.
                  *   Apache Ranger Support in Ozone.
                  *   Extensive failure testing for Ozone.
                The RC artifacts are available at https://home.apache.org/~ajay/ozone-0.4.0-alpha-rc2/
                The RC tag in git is ozone-0.4.0-alpha-RC2 (git hash 4ea602c1ee7b5e1a5560c6cbd096de4b140f776b)
                Please try out<https://cwiki.apache.org/confluence/display/HADOOP/Running+via+Apache+Release>,
vote, or just give us feedback.
                The vote will run for 5 days, ending on May 4, 2019, 04:00 UTC.
                Thank you very much,
            To unsubscribe, e-mail: hdfs-dev-unsubscribe@hadoop.apache.org
            For additional commands, e-mail: hdfs-dev-help@hadoop.apache.org
        To unsubscribe, e-mail: hdfs-dev-unsubscribe@hadoop.apache.org
        For additional commands, e-mail: hdfs-dev-help@hadoop.apache.org
View raw message