Re: RES: [GUSDEV] class notdefyet on building GUS 3.5

[Michael S. <msa...@pc...>]

Hi Fabricio,

I believe the dbiDsn should be:

dbiDsn=3Ddbi:Pg:dbname=3Dgus35

--Mike

On 7/4/05 11:35 AM, "Fabr=C3=ADcio" <fab...@de...> wrote:

> Hi Mike,
>=20
> I checked your suggestions and I noticed that we didn=E2=80=99t get SQL errors =
when
> running the build command with -installDBSchema. The core.tableinfo table
> has exactly 954 rows. I don=E2=80=99t known if this is the final row number; th=
e
> schema seems all ok, I really don=E2=80=99t know.
>=20
> The build output with -installDBSchema is here again:
>=20
> http://www.de9.ime.eb.br/~fabricio/out.txt
>=20
>=20
> About our jdbcDsn and dbiDsn we have this in the gus.properties file:
>=20
> dbiDsn=3Ddbi:Pg:gus35
> jdbcDsn=3Djdbc:postgresql://localhost/gus35
>=20
> Is it not right?
>=20
> I tested this jdbc connection string in a simple java class and I could
> connect and manipulate the database without problems.
>=20
> Can you have an idea about the problem? I can't visualize where we are
> faulting.=20
>=20
> Thanks a lot for your help!
>=20
> Fabr=C3=ADcio
>=20
> -----Mensagem original-----
> De: Michael Saffitz [mailto:msa...@pc...]
> Enviada em: segunda-feira, 4 de julho de 2005 01:52
> Para: Fabr =C3=AD cio; Gusdev gusdev-gusdev
> Assunto: Re: [GUSDEV] class notdefyet on building GUS 3.5
>=20
>=20
> Hi Fabricio,
>=20
> This is usually indicative that the schema was not properly loaded into t=
he
> RDBMS.  Can you review the output from the very first run (when you did
> -installDBSchema) to confirm there were no SQL errors?  Can you connect t=
o
> your DB and confirm that the objects exist-- especially that the
> core.tableinfo table is fully populated.
>=20
> If the DB was not fully installed, you must resolve the issue preventing
> installation, and then remove all GUS objects that may have been partiall=
y
> created (i.e. Users/schemas, tables, sequences, etc.)  Then you can try
> again.  (I believe this is outlined in the install docs).
>=20
> If all of the objects are being fully installed, then it may be that your
> jdbc connection string has different settings than your perl connection
> string.  Make sure that your $GUS_CONFIG_FILE has the correct dbiDsn for
> your database.
>=20
> Let me know how the above goes and we'll figure out the next step.
>=20
> --Mike
>=20
> On 7/3/05 10:18 PM, "Fabr=C3=ADcio" <fab...@de...> wrote:
>=20
>> Hello all,
>>=20
>> =20
>>=20
>> We=C2=B9re trying to install the new GUS 3.5 release, and we=C2=B9re having a
> problem
>> when we have to run =C2=B3build GUS install -append -installDBSchema=C2=B2. Afte=
r a
>> time running the building process stops whit many errors like this:
>>=20
>> =20
>>=20
>>=20
> [javac]/usr/local/GUS/GUS35/project_home/GUS/Model/src/java/org/gusdb/mod=
el/
>> Core/AlgorithmInvocation_Row.java:99: cannot resolve symbol
>> [javac]symbol  : class notdefyet
>> [javac]location: class org.gusdb.model.Core.AlgorithmInvocation_Row
>> [javac]public notdefyet getModificationDate () { return
>> (notdefyet)get_Attribute("modification_date"); }
>> [javac]            ^
>>=20
>> =20
>>=20
>> Despite these errors, the schema was generated in our postgres, but we
> don=C2=B9t
>> know if it=C2=B9s complete.
>>=20
>> =20
>>=20
>> We tried to build again using only the command =C2=B3build GUS install =C2=ADapp=
end=C2=B2
>> and we put the output in a file as you can see here:
>>=20
>> =20
>>=20
>> http://www.de9.ime.eb.br/~fabricio/out.txt
>>=20
>> =20
>>=20
>> Can you tell us why the notdefyet class isn=C2=B9t found?
>>=20
>> =20
>>=20
>> Thanks a lot,
>>=20
>> =20
>>=20
>> Fabr=C3=ADcio.
>>=20
>> =20
>>=20
>> =20
>>=20
>> =20
>>=20
>=20
>=20