[EMAIL PROTECTED] wrote:
Deepa Remesh <[EMAIL PROTECTED]> writes:
On 3/16/06, Dyre Tjeldvoll (JIRA) <[email protected]> wrote:
[ http://issues.apache.org/jira/browse/DERBY-1093?page=all ]
Dyre Tjeldvoll updated DERBY-1093:
----------------------------------
Attachment: upgrade.txt
Based on the information in DERBY-573 I have run the runphases script to test upgrade. But I don't
really understand the output (there's a lot of it) since it contains both "PASS" and
"FAIL" and exceptions... I have attached the output here, so hopefully someone can have a
look at it and tell me if there are any problems...
Currently, we have to compare the output to the master file
phaseTester.out. Unfortunately, there are a lot of existing diffs when
you run this test with the trunk. There is a diff file attached to
DERBY-1076 which shows the exisiting diffs. It would be great if you
can take a look at this. It may not be easy to look through these
diffs since there are quite a few. I will also try to take a look at
your output and post to the jira if I find anything.
OK, thanks for the explanation :) I just saw another mail describing
how the phaseTester master actually depends on the master for
metadata.java. And since I have updated this master (the
patch requires it) I cannot expect the phaseTester to run cleanly
(even modulo the diffs described in DERBY-1076)...
I have run the phaseTester on a clean build. The output from phaseTester
was not comparable with the phaseTester.out file, it contained many
additional failures.
I also found that phase tester gives different output dependent on which
version of Derby 10.1 you upgrade from.
Andreas
