Just for interest sake, are you going to record a row in the table for each log statement, or a row for each log file (as it were)?
Here at work we started doing a row per log statement, but our code produces a prodigious number of log statments, so that each run of the application was filling up about 100,000 rows. We found that creating a normal log file, and then inserted the whole thing as a blob into the db was much more efficient. I only wish I had more time in the day, this would be a fun project too :-( Richard On Wed, 2003-02-19 at 07:51, Ceki Gülcü wrote: > > Unless someone does it before, I intend to rewrite the JDBCAppender > completely. Removing the existing JDBCAppender is my preferred choice. > > At 22:11 18.02.2003 -0800, you wrote: > >I propose that we place the current version of > >o.a.log4j.jdbc.JDBCAppender.java into the log4j-sandbox (I don't know if we > >should remove it altogether from the v1.3 release?). In doing so, > >interested log4j developers need to step forward to begin the task of > >continuing its development. I know Kevin Steppe had more ideas for evolving > >this class, and others have spoken up as well. Sounds like it needs the > >ability to support different jdbc driver behaviors, clob's, etc. It seems > >that this appender is popular with the log4j user community, so I think it > >is right for it to be given some needed attention and upgrades. > > > >Interested developers, demonstrating interest, proposals, and ability can be > >granted committer rights to the sandbox cvs. > > > >Once JDBCAppender reaches a stable point of evolution, we can vote to > >re-admit it to the core release. > > > >+1 > > > >-Mark > > -- > Ceki > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
