On Sun, Sep 23, 2007 at 04:38:45AM -0400, Clint Adams wrote: > Looking in obj/ALL.OUT (instead of the grep) will give you more useful > information. The tests are all TCL scripts in the test/ directory. > > The detailed log and the scripts should help you isolate what's going on > better.
I'm going to expound on the log failures from db_4.6.21-2 and compare them with db4.5_4.5.20-9 which builds successfully. <http://buildd.debian.org/fetch.cgi?&pkg=db&ver=4.6.21-2&arch=m68k&stamp=1192597164&file=log> <http://buildd.debian.org/fetch.cgi?&pkg=db4.5&ver=4.5.20-9&arch=m68k&stamp=1191931936&file=log> The full ALL.OUT from db_4.6.21-2 at the following. <http://people.debian.org/~smarenka/tmp/db_4.6.21-2_ALL.OUT> * Error line 1 | FAIL: run_subsystem: env env007: env66: mmap: Cannot allocate memory This one occurs in db4.5. If I read it correctly, this fails because of insufficent resources. * Error lines 2 & 3 From db. | FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL:05:14:45 (00:00:00) db_get: expected DATA1, got | FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL:05:14:45 (00:00:00) ../dist/../test/envscript.tcl run: FAIL:05:14:45 (00:00:00) db_get: expected DATA1, got : pid 21249: expected 0, got 1 The next line reads | Env012.d: Second process cannot join without -recover after first process is killed. From db4.5. || db4.5: FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL:07:29:17 (00:00:00) db_get: expected DATA1, got || db4.5: FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL:07:29:17 (00:00:00) ../dist/../test/envscript.tcl run: pid 21801: expected 0, got 1 * Error lines 4 & 5 From db. | FAIL: error in ./TESTDIR/env012.log.p3 : ./TESTDIR/env012.log.p3:FAIL:05:15:36 (00:00:00) db_get: expected DATA1, got | FAIL: error in ./TESTDIR/env012.log.p3 : ./TESTDIR/env012.log.p3:FAIL:05:15:36 (00:00:00) ../dist/../test/envscript.tcl run: FAIL:05:15:36 (00:00:00) db_get: expected DATA1, got : pid 21409: expected 0, got 1 The next line reads | Env012.f: Empty slot shouldn't cause automatic recovery. From db4.5. || FAIL: error in ./TESTDIR/env012.log.p3 : ./TESTDIR/env012.log.p3:FAIL:07:30:11 (00:00:00) db_get: expected DATA1, got || FAIL: error in ./TESTDIR/env012.log.p3 : ./TESTDIR/env012.log.p3:FAIL:07:30:12 (00:00:01) ../dist/../test/envscript.tcl run: pid 21961: expected 0, got 1 So to this point we have the same FAILURES in both db and db4.5, excepting a whole string of the following in db4.5. || FAIL: Problem reported killing process 21915: child process exited abnormally So perhaps these are the errors of interest? They are all clustered together in the log. | FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL: The DB_RECOVER flag was not specified, and recovery is needed | FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL: opening env returned DB_RUNRECOVERY: Fatal error, run database recovery | FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL:05:15:59 (00:00:00) envopen: expected TRUE, got DB_RUNRECOVERY: Fatal error, run database recovery | FAIL: error in ./TESTDIR/env012.log.p2 : ./TESTDIR/env012.log.p2:FAIL:05:15:59 (00:00:00) ../dist/../test/envscript.tcl run: FAIL:05:15:59 (00:00:00) envopen: expected TRUE, got DB_RUNRECOVERY: Fatal error, run database recovery: pid 21459: expected 0, got 1 I have no idea what the implications of any of these errors are. If you can provide specific tests I can run or ideas to try, I'm game. I can also make sure you have access to a box or an aranym emulator if need be. Thanks, Stephen -- Stephen R. Marenka If life's not fun, you're not doing it right! <[EMAIL PROTECTED]>
signature.asc
Description: Digital signature
