Mailing-List: contact dev-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Date: Tue, 4 Feb 2014 23:55:09 +0000 (UTC)
From: "Brian Muray (JIRA)" <jira@apache.org>
To: hive-dev@hadoop.apache.org
Message-ID: <JIRA.12693045.1391497752700.38262.1391558109061@arcas>
In-Reply-To: <JIRA.12693045.1391497752700@arcas>
References: <JIRA.12693045.1391497752700@arcas>
Subject: [jira] [Commented] (HIVE-6363) IllegalArgumentException is thrown
 instead of SQLException
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


    [ https://issues.apache.org/jira/browse/HIVE-6363?page=3Dcom.atlassian.=
jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D13891=
462#comment-13891462 ]=20

Brian Muray commented on HIVE-6363:
-----------------------------------

A similar issue was fixed in HIVE-4194.  The changes were overwritten by th=
e changes to HIVE-4802.

> IllegalArgumentException is thrown instead of SQLException
> ----------------------------------------------------------
>
>                 Key: HIVE-6363
>                 URL: https://issues.apache.org/jira/browse/HIVE-6363
>             Project: Hive
>          Issue Type: Bug
>          Components: JDBC
>    Affects Versions: 0.10.0, 0.11.0, 0.12.0
>            Reporter: chandra sekhar gunturi
>            Priority: Critical
>              Labels: JDBC, unit-test
>         Attachments: HIVE-6363.patch.1.txt, HIVE-6363.patch.2.txt
>
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> parseURL in the following code is throwing IllegalArgumentException=20
> http://svn.apache.org/viewvc/hive/trunk/jdbc/src/java/org/apache/hive/jdb=
c/Utils.java?view=3Dmarkup=20
> This is going to break other JDBC based connectors because java.sql.Drive=
rManager doesn=E2=80=99t catch IllegalArgumentException while probing for c=
orrect Driver for a given URL.=20
> A simple test case can have class.forName(org.apache.hive.jdbc.HiveDriver=
) (Loading hiveserver2 JDBC driver) followed by class.forName(org.apache.ha=
doop.hive.jdbc.HiveDriver)(Loading hiveserver JDBC driver).
> In this case hiveserver connection will fail with BAD URL format for hive=
server. If you reverse the driver loading to hiveserver followed by hiveser=
ver2, both the connections will be successful.
> Following code in java.sql.DriverManager is causing the issue=20
> [[=20
> // Worker method called by the public getConnection() methods.=20
> private static Connection getConnection(=20
> // Walk through the loaded registeredDrivers attempting to make a connect=
ion.=20
> // Remember the first exception that gets raised so we can reraise it.=20
> for(DriverInfo aDriver : registeredDrivers) {=20
> // If the caller does not have permission to load the driver then=20
> // skip it.=20
> if(isDriverAllowed(aDriver.driver, callerCL)) {=20
> try {=20
> Connection con =3D aDriver.driver.connect(url, info);=20
> if (con !=3D null) {=20
> // Success!=20
> println("getConnection returning " + aDriver.driver.getClass().getName())=
;=20
> return (con);=20
> }=20
> } catch (SQLException ex) {=20
> if (reason =3D=3D null) {=20
> reason =3D ex;=20
> }=20
> }=20
> } else {=20
> println(" skipping: " + aDriver.getClass().getName());=20
> }=20
> }=20
> }=20
> ]]=20
> Marking it as critical because this is going to restrict consuming JDBC d=
river in production environment where many drivers are loaded on requiremen=
t rather than statically loading all drivers.


--
This message was sent by Atlassian JIRA
(v6.1.5#6160)