On Fri, 2005-04-01 at 14:37 -0500, D. Richard Hipp wrote: > On Fri, 2005-04-01 at 09:06 -0500, G. Roderick Singleton wrote: > > Running make test as a normal user, make test halts in wait() at the > > following test: > > > > lock-2.1... Ok > > lock-2.2... Ok > > lock-2.3.1... Ok > > lock-2.3.2... Ok > > lock-2.4.1... Ok > > lock-2.4.2... Ok > > lock-2.5... Ok > > lock-2.8... > > > > What is supposed to happen? Last time I had failed tests the solution > > was run as user. Now ... Do prior releases need to be removed from the > > system before building the new? > > > > The test that is "locking up" (lock-2.8) is a test of the > sqlite3_busy_timeout() API. It suppose to time out after > 400 milliseconds. It works fine on my SuSE 9.2 box. > > If you want to help debug this, recompile "testfixture" > with debugging symbols turned on. Then run testfixture > in gdb with argument ../sqlite/test/lock.test. Set a > breakpoint on sqliteDefaultBusyCallback(). Try to figure > out what is going wrong.
Between posting my message and now, I enabled debug in the Makefile and re-made sqlite. I also am re-running fulltest. -- G. Roderick Singleton <[EMAIL PROTECTED]> PATH tech
