Hi Ed, I took a look at the code, and I think (as you probably already have figured out) that the function guessSchemaVersion in SchemaVersion.java is not correctly guessing the version. This may also be messing up the getCVTerms function in JdbcChadoAdapter.java. I thought that I would rebuild the code to see if I could fix this, but 2 issues appear:
1. The code to download on sourceforge for gmod-1.1 is Apollo 1.11.1 rather that 1.11.6, the newest version. Is this code available? 2. There is no makefile in the src/java directory to rebuild as per the README. Any thoughts are appreciated. I will be happy to submit a fix if I figure one out. Thanks, Jared -----Original Message----- From: Ed Lee [mailto:e...@berkeleybop.org] Sent: Wednesday, June 09, 2010 12:48 PM To: Becksfort, Jared Cc: 'Scott Cain'; apollo@fruitfly.org Subject: RE: [apollo] Apollo hanging at 50% Hi Jared, I don't think that using the FlybaseChadoInstance would give you the pre-FMB schema I mentioned. But yes, you're better off using the RiceChadoInstance, which will use the standard Chado install (unless you're actually using a FB dump or derivative). You'll want to change the <clsName> element to: <clsName>apollo.dataadapter.chado.jdbc.RiceChadoInstance</clsName> rather than <clsName>apollo.dataadapter.chado.jdbc.RiceInstance</clsName> (note the "Chado" in RiceChadoInstance). Can you check that your database contains the "db" table? As I mentioned, it's used to determine the version of Chado, as the pre-FMB version did not contain that table. Cheers, Ed On Wed, 9 Jun 2010, Becksfort, Jared wrote: > Thanks, that is helpful. I'm sorry if I'm being dense, but how should I > configure it to look for the newer version? After looking at the Apollo > website, I tried changing in "referenceInstance": > > <clsName>apollo.dataadapter.chado.jdbc.FlybaseChadoInstance</clsName> > > To: > > <clsName>apollo.dataadapter.chado.jdbc.RiceInstance</clsName> > > But this resulted in: > > ERROR: ChadoAdapter.generateChadoInstance(): > java.lang.ClassNotFoundException: apollo.dataadapter.chado.jdbc.RiceInstance > > Which must mean that I am entering the wrong thing here or need to get this > file. > > Thanks, > Jared > > -----Original Message----- > From: Scott Cain [mailto:sc...@scottcain.net] > Sent: Tuesday, June 08, 2010 2:42 PM > To: Becksfort, Jared > Cc: Ed Lee; apollo@fruitfly.org > Subject: Re: [apollo] Apollo hanging at 50% > > Hi Ed, > > I think Jared's instance of Apollo is misconfigured to look for an old > version of Chado, no? > > Scott > > > On Tue, Jun 8, 2010 at 3:03 PM, Becksfort, Jared > <jared.becksf...@stjude.org> wrote: >> Ed, >> >> Thank you for replying. I suppose I wasn't as clear as I could have been. >> The Chado schema (gmod-1.1) that I am using definitely has the cvterm table, >> and I can access it from the command line and from a remote system using >> PgAdmin III. It has about 36k rows in it, probably from loading ontologies >> during setup. >> >> My guess initial guess was that it is something that I didn't do correctly >> in the XML files, and this is still a possibility. I've attached it in case >> anyone was curious enough to take a stab at it (search for 'pcgp'). I also >> thought that perhaps Apollo was not able to use the newer version of the >> Chado schema. I don't think that the schema is faulty because I have used a >> lot of gmod tools to load it with data. >> >> My concern about the query: >> SELECT cv.cvterm_id, cv.name FROM cvterm cv, cv c WHERE c.cv_id = >> cv.cv_id AND c.cvname='sequence' >> >> Comes from when I run it using PgAdmin rather than Apollo. I get this error: >> >> ERROR: column c.cvname does not exist >> LINE 1: ...FROM cvterm cv, cv c WHERE c.cv_id = cv.cv_id AND >> c.cvname='... >> ^ >> ********** Error ********** >> >> ERROR: column c.cvname does not exist >> SQL state: 42703 >> Character: 80 >> >> My version of Chado has cv.name instead of cv.cvname (notice the table >> aliases in the query). Granted, this could be the next problem that I >> encounter rather than the one I am encountering now. I just wondered if >> they were in the same try-catch block or whatever Java uses (not a Java guy, >> sorry), then maybe the error message was giving the wrong impression. It >> could be an error that is preventing DB connectivity, but in that case, I >> would have expected a different error, after: >> >> INFO : connecting to database pcgp as user postgres at >> jdbc:postgresql://ergatisvm:5432 >> >> Thanks for your help! >> Jared >> >> -----Original Message----- >> From: Ed Lee [mailto:e...@berkeleybop.org] >> Sent: Tuesday, June 08, 2010 12:31 PM >> To: Becksfort, Jared >> Cc: 'apollo@fruitfly.org' >> Subject: Re: [apollo] Apollo hanging at 50% >> >> Hi Jared, >> >> Sounds like your schema might be faulty somehow. "cv.cvname" comes from >> an older, pre-Flybase migration build of Chado. The current schema >> should have it as "cv.name". Apollo supports both schemas and it >> figures out which schema to use by querying the "db" table, which >> the pre-FMB version lacks. So it sounds like you're missing that >> table. Also, the PostgreSQL exception is complaining about the >> "cvterm" table not existing. I wouldn't expect PostgreSQL to >> complain about a table not existing when it's due to a non existing >> column. >> >> Cheers, >> Ed >> >> On Mon, 7 Jun 2010, Becksfort, Jared wrote: >> >>> >>> Hello, >>> >>> >>> >>> I am trying to view a Chado installation using Apollo. After I have >>> selected the database to view, Apollo will start to load it but then hang at >>> the 50% mark. >>> >>> >>> >>> I have installed GMOD-1.1 with all of the ontologies. >>> >>> >>> >>> Here is the relevant output in the logs: >>> >>> >>> >>> INFO : APOLLO_ROOT: >>> /nfs_exports/apps/gnu-apps/ergatis/software/apollo-1.11.5/Apollo >>> >>> INFO : OS name: Linux >>> >>> INFO : OS arch: amd64 >>> >>> INFO : Java version: 1.6.0_15 >>> >>> INFO : reading config >>> file/nfs_exports/apps/gnu-apps/ergatis/software/apollo-1.11.5/Apollo/conf/apoll >>> o.cfg >>> >>> INFO : using history file /home/ergatis/.apollo/apollo.history >>> >>> INFO : using backup file /home/ergatis/.apollo/apollo.backup >>> >>> INFO : this is Apollo Genome Annotation and Curation Tool, version 1.11.5, >>> last updated Mar 16 2010 >>> >>> INFO : connecting to database pcgp as user postgres at >>> jdbc:postgresql://ergatisvm:5432 >>> >>> ERROR: SQLException running SELECT cv.cvterm_id, cv.name FROM cvterm cv, cv >>> c WHERE c.cv_id = cv.cv_id AND c.cvname='sequence' >>> >>> org.postgresql.util.PSQLException: ERROR: relation "cvterm" does not exist >>> >>> >>> >>> at >>> org.postgresql.util.PSQLException.parseServerError(PSQLException.java:139) >>> >>> >>> >>> I can connect to the database from the command line. I also notice that >>> even if the database were properly connected, this query will fail in >>> gmod-1.1 because "cvname" does not exist in the cvterm table - it should >>> just be name. It makes me wonder if that is the actual error but it is >>> being caught by an exception to the whole query. >>> >>> >>> >>> Any help will be appreciated. Please let me know if you need anything else >>> from me. >>> >>> >>> >>> Thanks! >>> >>> Jared Becksfort >>> >>> >>> ____________________________________________________________________________ >>> Email Disclaimer: www.stjude.org/emaildisclaimer >>> >>> >> >> _______________________________________________ >> apollo mailing list >> apollo@fruitfly.org >> http://mail.fruitfly.org/mailman/listinfo/apollo >> >> > > > > -- > ------------------------------------------------------------------------ > Scott Cain, Ph. D. scott at scottcain dot > net > GMOD Coordinator (http://gmod.org/) 216-392-3087 > Ontario Institute for Cancer Research > > > > _______________________________________________ apollo mailing list apollo@fruitfly.org http://mail.fruitfly.org/mailman/listinfo/apollo
