pgsql: Remove prepared transactions from main isolation test schedule.
Remove prepared transactions from main isolation test schedule.
There is no point in running this test when prepared transactions are disabled,
which is the default. New make targets that include the test are provided. This
will save some useless waste of cycles on buildfarm machines.
Backpatch to 9.1 where these tests were introduced.
Branch
------
master
Details
-------
http://git.postgresql.org/pg/commitdiff/ae55d9fbe3871a5e6309d9b91629f1b0ff2b8cba
Modified Files
--------------
src/test/isolation/Makefile | 10 ++++++++++
src/test/isolation/isolation_schedule | 1 -
2 files changed, 10 insertions(+), 1 deletions(-)
Andrew Dunstan <andrew@dunslane.net> writes:
Remove prepared transactions from main isolation test schedule.
There is no point in running this test when prepared transactions are disabled,
which is the default. New make targets that include the test are provided. This
will save some useless waste of cycles on buildfarm machines.
Having done that, shouldn't we remove prepared-transactions_1.out (the
"expected" file for the prepared-transactions-disabled case)?
Having a megabyte-sized test file for that case has always seemed pretty
wasteful to me. Now that the test won't be run unless intentionally
selected, it seems like people who are using it would expect it to
actually test prepared transactions --- so having it "pass" when they're
disabled seems wrong.
regards, tom lane
On 07/20/2012 04:17 PM, Tom Lane wrote:
Andrew Dunstan <andrew@dunslane.net> writes:
Remove prepared transactions from main isolation test schedule.
There is no point in running this test when prepared transactions are disabled,
which is the default. New make targets that include the test are provided. This
will save some useless waste of cycles on buildfarm machines.Having done that, shouldn't we remove prepared-transactions_1.out (the
"expected" file for the prepared-transactions-disabled case)?Having a megabyte-sized test file for that case has always seemed pretty
wasteful to me. Now that the test won't be run unless intentionally
selected, it seems like people who are using it would expect it to
actually test prepared transactions --- so having it "pass" when they're
disabled seems wrong.
Good point. Will do.
cheers
andrew