Mailing-List: contact users-help@jackrabbit.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@jackrabbit.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
MIME-Version: 1.0
In-Reply-To: <AANLkTik33jtSV-qULUhDA5++oJ8Eq5VG6JXPpyK8nn9n@mail.gmail.com>
References: <AANLkTin_SvpTEAn9Ezg_wDgSMhWWTJk-sa=-Y9tZwMjO@mail.gmail.com>
 <9C50F30A-40CD-4DE9-9E4A-03CE0383CBB0@gmail.com>
 <AANLkTinh_WwjzRPjV9hG6tyE+YES+-B+Zh0mtpzNq+4h@mail.gmail.com>
 <C4D6AAFD-F5FF-4247-A523-024BF117BE18@gmail.com>
 <AANLkTikzWBDFaQRx1uNv1TUXk_xpWK58LBmg6uXswXvD@mail.gmail.com>
 <E554BB7B-9D3A-49BD-BBFE-B28D89F31A46@gmail.com>
 <AANLkTikDZY6y+DeO5HP3Qb24GNBOCG4j9AbvyxLGxNb0@mail.gmail.com>
 <AANLkTimHt5P+c9jC-Gv0oVRw4DLkpguAqhMkuStK-QbS@mail.gmail.com>
 <AANLkTin9AddqXPCbp_A0Jp6C_h8mwh-UTebWqMAddCZQ@mail.gmail.com>
 <AANLkTimW+DTSG2XnAW18UbO43aP3F7C6y4s_OSOk1vXK@mail.gmail.com>
 <AANLkTikgHPqCgFkhLZ40V_Q36i2hUBJy1qJ-U4bfKmYw@mail.gmail.com>
 <AANLkTi=R8=vD+i38joVQ_6qVsn_Bf46pHHmj3+9Pkzmn@mail.gmail.com>
 <AANLkTik33jtSV-qULUhDA5++oJ8Eq5VG6JXPpyK8nn9n@mail.gmail.com>
From: Carl Hall <carl@hallwaytech.com>
Date: Mon, 16 Aug 2010 13:08:13 -0400
Message-ID: <AANLkTikg78xfsscYqVkkvQ6mM8cdCq=vJc1houhNrFLQ@mail.gmail.com>
Subject: Re: Clustering
To: users@jackrabbit.apache.org
Content-Type: multipart/alternative; boundary=0016364d2335e49825048df3e1b0

--0016364d2335e49825048df3e1b0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

And the final change to fix my issue: set databaseType=3D"oracle" in my
journal definition.

OraclePersistenceManager does this without me setting the parameter
explicitly (~line 97).  Should OracleDatabaseJournal assume the defaults
that OraclePersistenceManager does?


On Mon, Aug 16, 2010 at 11:25 AM, Carl Hall <carl@hallwaytech.com> wrote:

> I started Sling and debugged into the JackRabbit code to see what SQL is
> failing.  This is what I've found.
>
> This SQL is run but fails with the 'invalid datatype' exception:
> create table JOURNAL (REVISION_ID BIGINT NOT NULL, JOURNAL_ID varchar(255=
),
> PRODUCER_ID varchar(255), REVISION_DATA varbinary)
>
> When I modify it in memory to this, it works:
> create table JOURNAL (REVISION_ID number(20,0) NOT NULL, JOURNAL_ID
> varchar(255), PRODUCER_ID varchar(255), REVISION_DATA blob)
>
> but then I find that this statement fails next:
> create table LOCAL_REVISIONS (JOURNAL_ID varchar(255) NOT NULL, REVISION_=
ID
> BIGINT NOT NULL)
>
> Same changes as above (BIGINT -> number(20,0)) and we're clear again.
>
> Should I file a JIRA for this?  If someone can give me direction, I don't
> mind making a patch or changing configuration where needed.
>
>
> On Mon, Aug 16, 2010 at 9:20 AM, Carl Hall <carl@hallwaytech.com> wrote:
>
>> Aside from the first 3 drop statements failing (tables didn't exist), th=
is
>> ran correctly.  I am changing my repository FS to be a DB FS and will tr=
y
>> again.  Maybe this will help in someway.
>>
>>
>> On Mon, Aug 16, 2010 at 9:05 AM, Thomas M=FCller <thomas.mueller@day.com=
>wrote:
>>
>>> Hi,
>>>
>>> > Oracle 9.
>>>
>>> I'm not sure what the problem might be... Could you run the following
>>> script to find out where exactly the problem is?
>>>
>>> drop table JOURNAL;
>>> drop table GLOBAL_REVISION;
>>> drop table LOCAL_REVISIONS;
>>> create table JOURNAL (REVISION_ID number(20,0) NOT NULL, JOURNAL_ID
>>> varchar(255), PRODUCER_ID varchar(255), REVISION_DATA blob);
>>> create unique index JOURNAL_IDX on JOURNAL (REVISION_ID);
>>> create table GLOBAL_REVISION (REVISION_ID number(20,0) NOT NULL);
>>> create unique index GLOBAL_REVISION_IDX on GLOBAL_REVISION (REVISION_ID=
);
>>> create table LOCAL_REVISIONS (JOURNAL_ID varchar(255) NOT NULL,
>>> REVISION_ID number(20,0) NOT NULL);
>>> insert into GLOBAL_REVISION VALUES(0);
>>> drop table JOURNAL;
>>> drop table GLOBAL_REVISION;
>>> drop table LOCAL_REVISIONS;
>>>
>>> If this works, then I don't know where the problem might be.
>>>
>>> Regards,
>>> Thomas
>>>
>>
>>
>

--0016364d2335e49825048df3e1b0--