You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@drill.apache.org by Vova Vysotskyi <vv...@gmail.com> on 2020/04/01 17:35:16 UTC
DRILL-7671: Verify Drill with cdh and hdp profiles
Hi all,
Some time ago I have noticed that Apache Drill has *cdh* and *hdp*
profiles, but it is unable to build when any of them is enabled.
I have created DRILL-7671 <https://issues.apache.org/jira/browse/DRILL-7671> to
fix it. One of the major unresolved issues I faced up with is verification
that Drill works correctly on both distributives.
Unfortunately, I don't have a cluster with any of those distributives and
corresponding sandboxes contain slightly outdated versions of ZooKeeper and
other projects to ensure that there are no issues.
Could someone who has access to cluster with any of these distributives,
help with verification of this fix, so we will be sure that there is
nothing to do before merging it to master.
Here is the branch with changes: DRILL-7671
<https://github.com/vvysotskyi/drill/tree/DRILL-7671>.
To verify the fix, please follow the next steps:
*git clone git@github.com:vvysotskyi/drill.git*
*git checkout DRILL-7671*
For Cloudera distributive, please build Drill with the *cdh* profile:
*mvn clean install -DskipTests -Pcdh*
For Hortonworks distributive, please build Drill with the *hdp* profile:
*mvn clean install -DskipTests -Phdp*
I think it would be good to do the following sanity checks:
- Configure and run Drill in distributed mode;
- Query some files from HDFS;
- If possible, configure Hive plugin and query Hive tables;
- If possible, configure HBase plugin and query HBase tables;
- If possible, configure any of JDBC clients and query Drill;
- Something else.
The branch above uses the following versions (latest available in
corresponding repositories):
*cdh* profile:
hadoop 3.1.1.7.0.3.0-79
hbase 2.2.0.7.0.3.0-79
hive 3.1.2000.7.0.3.0-79
zookeeper 3.5.5.7.0.3.0-79
*hdp* profile:
hadoop 3.1.1.3.1.5.6-1
hbase 2.0.2.3.1.0.6-1
hive 3.1.0.3.1.0.6-1
zookeeper 3.4.6.3.3.1.6-1
Kind regards,
Volodymyr Vysotskyi
Re: DRILL-7671: Verify Drill with cdh and hdp profiles
Posted by Vova Vysotskyi <vv...@gmail.com>.
Hi all,
Looks like no one found any issues.
I'll proceed with creating a pull request in this case.
Kind regards,
Volodymyr Vysotskyi
On Wed, Apr 1, 2020 at 8:35 PM Vova Vysotskyi <vv...@gmail.com> wrote:
> Hi all,
>
> Some time ago I have noticed that Apache Drill has *cdh* and *hdp*
> profiles, but it is unable to build when any of them is enabled.
>
> I have created DRILL-7671
> <https://issues.apache.org/jira/browse/DRILL-7671> to fix it. One of the
> major unresolved issues I faced up with is verification that Drill works
> correctly on both distributives.
> Unfortunately, I don't have a cluster with any of those distributives and
> corresponding sandboxes contain slightly outdated versions of ZooKeeper and
> other projects to ensure that there are no issues.
>
> Could someone who has access to cluster with any of these distributives,
> help with verification of this fix, so we will be sure that there is
> nothing to do before merging it to master.
>
> Here is the branch with changes: DRILL-7671
> <https://github.com/vvysotskyi/drill/tree/DRILL-7671>.
>
> To verify the fix, please follow the next steps:
>
> *git clone git@github.com:vvysotskyi/drill.git*
> *git checkout DRILL-7671*
>
> For Cloudera distributive, please build Drill with the *cdh* profile:
> *mvn clean install -DskipTests -Pcdh*
>
> For Hortonworks distributive, please build Drill with the *hdp* profile:
> *mvn clean install -DskipTests -Phdp*
>
> I think it would be good to do the following sanity checks:
> - Configure and run Drill in distributed mode;
> - Query some files from HDFS;
> - If possible, configure Hive plugin and query Hive tables;
> - If possible, configure HBase plugin and query HBase tables;
> - If possible, configure any of JDBC clients and query Drill;
> - Something else.
>
> The branch above uses the following versions (latest available in
> corresponding repositories):
>
> *cdh* profile:
> hadoop 3.1.1.7.0.3.0-79
> hbase 2.2.0.7.0.3.0-79
> hive 3.1.2000.7.0.3.0-79
> zookeeper 3.5.5.7.0.3.0-79
>
> *hdp* profile:
> hadoop 3.1.1.3.1.5.6-1
> hbase 2.0.2.3.1.0.6-1
> hive 3.1.0.3.1.0.6-1
> zookeeper 3.4.6.3.3.1.6-1
>
> Kind regards,
> Volodymyr Vysotskyi
>
Re: DRILL-7671: Verify Drill with cdh and hdp profiles
Posted by Vova Vysotskyi <vv...@gmail.com>.
Hi all,
Looks like no one found any issues.
I'll proceed with creating a pull request in this case.
Kind regards,
Volodymyr Vysotskyi
On Wed, Apr 1, 2020 at 8:35 PM Vova Vysotskyi <vv...@gmail.com> wrote:
> Hi all,
>
> Some time ago I have noticed that Apache Drill has *cdh* and *hdp*
> profiles, but it is unable to build when any of them is enabled.
>
> I have created DRILL-7671
> <https://issues.apache.org/jira/browse/DRILL-7671> to fix it. One of the
> major unresolved issues I faced up with is verification that Drill works
> correctly on both distributives.
> Unfortunately, I don't have a cluster with any of those distributives and
> corresponding sandboxes contain slightly outdated versions of ZooKeeper and
> other projects to ensure that there are no issues.
>
> Could someone who has access to cluster with any of these distributives,
> help with verification of this fix, so we will be sure that there is
> nothing to do before merging it to master.
>
> Here is the branch with changes: DRILL-7671
> <https://github.com/vvysotskyi/drill/tree/DRILL-7671>.
>
> To verify the fix, please follow the next steps:
>
> *git clone git@github.com:vvysotskyi/drill.git*
> *git checkout DRILL-7671*
>
> For Cloudera distributive, please build Drill with the *cdh* profile:
> *mvn clean install -DskipTests -Pcdh*
>
> For Hortonworks distributive, please build Drill with the *hdp* profile:
> *mvn clean install -DskipTests -Phdp*
>
> I think it would be good to do the following sanity checks:
> - Configure and run Drill in distributed mode;
> - Query some files from HDFS;
> - If possible, configure Hive plugin and query Hive tables;
> - If possible, configure HBase plugin and query HBase tables;
> - If possible, configure any of JDBC clients and query Drill;
> - Something else.
>
> The branch above uses the following versions (latest available in
> corresponding repositories):
>
> *cdh* profile:
> hadoop 3.1.1.7.0.3.0-79
> hbase 2.2.0.7.0.3.0-79
> hive 3.1.2000.7.0.3.0-79
> zookeeper 3.5.5.7.0.3.0-79
>
> *hdp* profile:
> hadoop 3.1.1.3.1.5.6-1
> hbase 2.0.2.3.1.0.6-1
> hive 3.1.0.3.1.0.6-1
> zookeeper 3.4.6.3.3.1.6-1
>
> Kind regards,
> Volodymyr Vysotskyi
>