Thread: XML regression test failure
$ make -C src/test/regress runtest [ ... ] test xml ... FAILED test stats ... ok test tablespace ... ok ======================== 1 of 105 tests failed. ======================== The regression.diffs are attached. Note that this reproduces consistently, but only occurs if I use the "runtest" Makefile target (i.e. run the regression tests in serial mode); "make check" works fine, for example. -Neil
Attachment
Neil Conway wrote: > The regression.diffs are attached. Note that this reproduces > consistently, but only occurs if I use the "runtest" Makefile target > (i.e. run the regression tests in serial mode); "make check" works > fine, for example. You need to re-initdb. -- Peter Eisentraut http://developer.postgresql.org/~petere/
Peter Eisentraut <peter_e@gmx.net> writes:
> Neil Conway wrote:
>> The regression.diffs are attached. Note that this reproduces
>> consistently, but only occurs if I use the "runtest" Makefile target
>> (i.e. run the regression tests in serial mode); "make check" works
>> fine, for example.
> You need to re-initdb.
IOW someone forgot to bump catversion when committing a change that
invalidated the regression tests? Preventing this type of confusion
is exactly what we invented catversion for.
regards, tom lane
Tom Lane wrote: > Peter Eisentraut <peter_e@gmx.net> writes: > > Neil Conway wrote: > >> The regression.diffs are attached. Note that this reproduces > >> consistently, but only occurs if I use the "runtest" Makefile > >> target (i.e. run the regression tests in serial mode); "make > >> check" works fine, for example. > > > > You need to re-initdb. > > IOW someone forgot to bump catversion when committing a change that > invalidated the regression tests? No. I should have said you need to reinstall, when you run the test against an already installed instance. Once reinstalled, the catversion check will trigger. -- Peter Eisentraut http://developer.postgresql.org/~petere/
On Sun, 2007-01-21 at 08:36 +0100, Peter Eisentraut wrote: > Tom Lane wrote: > > Peter Eisentraut <peter_e@gmx.net> writes: > > > Neil Conway wrote: > > >> The regression.diffs are attached. Note that this reproduces > > >> consistently, but only occurs if I use the "runtest" Makefile > > >> target (i.e. run the regression tests in serial mode); "make > > >> check" works fine, for example. > > > > > > You need to re-initdb. > > > > IOW someone forgot to bump catversion when committing a change that > > invalidated the regression tests? > > No. I should have said you need to reinstall, when you run the test > against an already installed instance. Once reinstalled, the > catversion check will trigger. The main point is the same as the catversion check however. There should be a message that says "reinstall to correct this", just like the catversion check does, or at very least a hint to say "have you tried a reinstall to correct this?" -- Simon Riggs EnterpriseDB http://www.enterprisedb.com