On 4 Sep 2012, at 7:14pm, esum <eric.b....@lmco.com> wrote: > I am getting odd results when I do a simple test such as doing a couple of > inserts to different tables from 4 threads. Some sql statements appear to > be printed twice or more when they should only be printed once. The sql > statements that are replicated vary from test to test, and I am lead to > believe that there may be some race condition. Is sqlite3_trace () > threadsafe?
As far as we know sqlite3_trace() is 100% threadsafe unless you have chosen options explained in http://www.sqlite.org/threadsafe.html which defeat thread safety for other parts of SQLite. > Is there a reason that the same sql statement is being passed > twice to the callback function when it is really only executed once? As a > note, the database itself appears to be perfectly fine with everything in > tact and as expected. Nice analysis. Thanks for posting enough that we understand your setup and can help you. I think you're being caught by something you didn't notice in the documentation: <http://www.sqlite.org/c3ref/profile.html> "The sqlite3_trace() callback is invoked with a UTF-8 rendering of the SQL statement text as the statement first begins executing. Additional sqlite3_trace() callbacks might occur as each triggered subprogram is entered." Notice the 'might'. What I am guessing is happening is that sometimes sqlite3_trace() is called, but is eventually held up because of access contention, a situation where it can't go ahead because the database is locked. Eventually the database becomes free again and this thread can go ahead, at which point _trace() gets called again. You might be able to play with your value for sqlite3_busy_timeout() and prove or disprove this. Whether I guessed right or not, the 'additional' suggests that there's no problem with your code, and you're just seeing some inner workings of SQLite you hadn't expected. Simon. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
