Stephen,
Stephen Ince wrote:
Werner,
Thanx for your respone.
Also how do you turn on SQL tracing with castor?
I was looking for the sql statements that castor executes
Oops, sorry, I thought you were referring to tracing of SQL statements
as executed within the RDBMS.
Can this be turned via a castor property?
No.
Or you have to set this via castor logging api?
Yes and no. Basically, everything is in place, as the relevant classes
emit log statements. You just have to configure your logger to the
appropriate log level (i.e. debug in this case). What logger are you
using ?
Dead link. "this message".
http://www.castor.org/jdo-faq.html#Castor-&-Logging
Well, well ... that statement is partly wrong (and thus should be
changed as well). On the JDO side of things, the code uses
commons-logging to emit log statements. All you have to do is to
configure your logger according to your needs. Have e.g. a look at the
log4j.xml in cpactf/src/test/resources to see how this is done.
Identity key sequence seems to be just as fast as HIGH/LOW.
Steve
----- Original Message ----- From: "Werner Guttmann"
<[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Thursday, February 07, 2008 1:07 PM
Subject: Re: [castor-user] key generator
Stephen,
let me address one issue at a time, and where I have go something to
say at all.
Werner
Stephen Ince wrote:
I am looking for some guidance on key generators with castor.
Basically pros/cons for speed and concurrency. Currently I am using
HIGH/LOW but I run into issues when the application has been running
for over a week. I get duplicate identity exceptions. I basically
have to restart the webapp.
1) HIGH/LOW -- the fastest?
2) IDENTITY -- is this as fast as HIGH/LOW?
3) SEQUENCE -- this seems to be problematic if you already have a
table populated. Would you need to have custom sql to update the
sequences?
It depends. If it's just a range between 1 and a certina value that is
assigend already, you could always initialize the sequence to a value
outside of Castor before 'turning on' your application (the one that
uses Castor for key generation). It gets far more complicated if the
values assigned already are equally (evently) distributed across the
complete value space, if not impossible.
Also how do you turn on SQL tracing with castor?
Hmm, if SQL tracing can be turned on by the means of a JDBC property,
you can pass JDBC properties ... in the JDO configuration file, afair.
I looked at the FAQ but
it pointed to a dead link.
Where is that link ?
Steve
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
