> On 12 Nov 2021, at 16:21, Tom Lane wrote:
>
> Daniel Gustafsson writes:
>> Attached is a small addition mentioning PG_TEST_NOCLEAN
>
> Maybe could use a bit of copy-editing, say
>
> Data directories will also be left behind for analysis when a test fails;
> they are named according to the te
Daniel Gustafsson writes:
> Attached is a small addition mentioning PG_TEST_NOCLEAN
Maybe could use a bit of copy-editing, say
Data directories will also be left behind for analysis when a test fails;
they are named according to the test filename. But if the environment
variable PG_TEST_NOCLEAN
> On 31 Oct 2021, at 23:13, Tom Lane wrote:
> Done that way; feel free to add more material to perl/README.
Attached is a small addition mentioning PG_TEST_NOCLEAN
--
Daniel Gustafsson https://vmware.com/
readme_retaindir.diff
Description: Binary data
The patch looks great. Thanks!
Kevin
On Sun, Oct 31, 2021 at 3:13 PM Tom Lane wrote:
> Daniel Gustafsson writes:
> > On 31 Oct 2021, at 19:28, Tom Lane wrote:
> >> Here's a quick stab at rearranging src/test/perl/README so that the
> >> initial section is all how-to-run-the-tests info, and ad
Daniel Gustafsson writes:
> On 31 Oct 2021, at 19:28, Tom Lane wrote:
>> Here's a quick stab at rearranging src/test/perl/README so that the
>> initial section is all how-to-run-the-tests info, and advice about
>> writing new tests starts after that. Your point about PG_TEST_NOCLEAN
>> could be
> On 31 Oct 2021, at 19:28, Tom Lane wrote:
>
> I wrote:
>> Daniel Gustafsson writes:
>>> Wouldn't it make more sense to start collecting troubleshooting advice in
>>> src/test/perl/README and instead refer to that in the boilerplate? I notice
>>> that we don't document for example PG_TEST_NOCL
I wrote:
> Daniel Gustafsson writes:
>> Wouldn't it make more sense to start collecting troubleshooting advice in
>> src/test/perl/README and instead refer to that in the boilerplate? I notice
>> that we don't document for example PG_TEST_NOCLEAN anywhere (which admittedly
>> is my fault), a trub
Daniel Gustafsson writes:
> Wouldn't it make more sense to start collecting troubleshooting advice in
> src/test/perl/README and instead refer to that in the boilerplate? I notice
> that we don't document for example PG_TEST_NOCLEAN anywhere (which admittedly
> is my fault), a trubleshooting sect
> On 30 Oct 2021, at 20:12, Tom Lane wrote:
> I'd be inclined to add just one sentence to the boilerplate text these use,
> along the lines of
> "If the tests fail, examining the logs left behind in tmp_check/log/
> may be helpful."
Wouldn't it make more sense to start collecting troubleshoo
Kevin Burke writes:
> I probably would not have looked there honestly; I was working in the
> terminal and had the source code right there.
Yeah, that was kind of what I thought.
> "For more information on Postgres's TAP tests, see the docs:
> https://www.postgresql.org/docs/devel/regress-tap.ht
I probably would not have looked there honestly; I was working in the
terminal and had the source code right there. Could we put a link to that
page in the README?
"For more information on Postgres's TAP tests, see the docs:
https://www.postgresql.org/docs/devel/regress-tap.html";
Kevin
On Sat,
Kevin Burke writes:
> I've been trying to run the SSL tests against my branch and that was
> tougher than expected because I didn't realize that additional output was
> being saved that I couldn't see - it wasn't even getting to the part where
> it could run the tests. This patch adds a note to th
12 matches
Mail list logo
