incubator-esme-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ethan Jewett <esjew...@gmail.com>
Subject Re: Search error when using JDBC for search (was "Metadata handling")
Date Thu, 22 Jul 2010 06:51:12 GMT
Not in front of my computer now, so please forgive mistakes.

My testing indicated that in the Compass object in Boot.scala the
conf.buildCompass() method call was failing silently during the setup
of the object.

The exception then occurs later on when the system attempts to use the
Compass.compass value, which I believe was never populated.

So, the actual failure occurs in the SearchMgr object, I think, so it
has Lift methods in the stack trace, but I don't think it's a Lift
issue. I'm not sure why we don't see an exception bubble up from the
buildCompass() method.

That is unfortunately about as far as far as I got.

Ethan

On Thursday, July 22, 2010, Imtiaz Ahmed H E <in.imtiaz@gmail.com> wrote:
> I mean, is this a compass thing ? but why is the exception name not printed in the log...
>
> ----- Original Message ----- From: "Imtiaz Ahmed H E" <in.imtiaz@gmail.com>
> To: <esme-dev@incubator.apache.org>
> Sent: Thursday, July 22, 2010 11:39 AM
> Subject: Re: Search error when using JDBC for search (was "Metadata handling")
>
>
>
> So that  I can take a quick step forward, can you tell me, how is the following trace
produced...is this a slf4j produced log message.
>
> Andwhy is the exception name itself, I mean , what it is, not printed along with the
stack trace, how can I get it printed.
>
> Enclosing the entire for expression in the 'search'  method in Message.scala in a try/catch
expression doesn't seem to result in the exception being caught by it. Apparently caught earlier,
how ? Is this a Lift thing...?
>
> ERROR - Array(org.apache.esme.model.Message$.search(Message.scala:152), org.apac
> he.esme.lib.SearchMgr$$anonfun$displaySearch$1$$anonfun$apply$1.apply(SearchMgr.
> scala:69), org.apache.esme.lib.SearchMgr$$anonfun$displaySearch$1$$anonfun$apply
> $1.apply(SearchMgr.scala:66), net.liftweb.common.Full.map(Box.scala:398), org.ap
> ache.esme.lib.SearchMgr$$anonfun$displaySearch$1.apply(SearchMgr.scala:66), org.
> apache.esme.lib.SearchMgr$$anonfun$displaySearch$1.apply(SearchMgr.scala:65), ne
>
>
> ----- Original Message ----- From: "Ethan Jewett" <esjewett@gmail.com>
> To: <esme-dev@incubator.apache.org>
> Sent: Wednesday, July 21, 2010 8:43 PM
> Subject: Search error when using JDBC for search (was "Metadata handling")
>
>
>
> To recreate the problem,
>
> 1. Replace the contents of /src/main/resources/props/compass.cfg.xml
> with the contents of /src/main/resources/props/compass.jndi.cfg.xml
> 2. mvn jetty:run
> 3. Point your web browser to http://localhost:8080
> 4. Log in if necessary
> 5. Do a search (it won't work and will just redirect you back to the main page)
> 6. You should now have a stack trace on the console
>
> Ethan
>
> On Wed, Jul 21, 2010 at 4:48 PM, Imtiaz Ahmed H E <in.imtiaz@gmail.com> wrote:
>
>
> <<<<<<<<<<<<
>
>
> You don't happen to have any JDBC/JNDI and/or Lucene experience do
> you? I'm totally stumped on the issue with search when we have it run
> using the database instead of the filesystem (ESME-205).
>
>
>
>
> No, no experience on those. Except have used JDBC for JavaDB access, in my
> last job and less than that of it with Oracle 8 years back.
>
> I just tried out a search which had a result of one match, on my system, and
> it worked fine with no exceptions.
>
> If I could reproduce the bug, I could comment on whether I should be
> assigned the Jira ticket...saw the exception trace in the ticket, 205, and
> the exception seems like a Scala coding problem...
>
> Imtiaz
>
> ----- Original Message ----- From: "Ethan Jewett" <esjewett@gmail.com>
> To: <esme-dev@incubator.apache.org>
> Sent: Wednesday, July 21, 2010 7:51 PM
> Subject: Re: Metadata handling (was "Release planning")
>
>
>
> Sounds good to me, at least until we can figure out what the next
> step/problem is on the metadata front. Just ping the list when you've
> uploaded the patch.
>
> You don't happen to have any JDBC/JNDI and/or Lucene experien

Mime
View raw message