Re: In-placre persistance change of a relation

On 2021-12-23 15:33:35 +0900, Kyotaro Horiguchi wrote:
> At Thu, 23 Dec 2021 15:01:41 +0900 (JST), Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com> wrote in
> > I added TAP test to excecise the in-place persistence change.
>
> We don't need a base table for every index. TAP test revised.

The tap tests seems to fail on all platforms. See
https://cirrus-ci.com/build/4911549314760704

E.g. the linux failure is

[16:45:15.569]
[16:45:15.569] # Failed test 'inserted'
[16:45:15.569] # at t/027_persistence_change.pl line 121.
[16:45:15.569] # Looks like you failed 1 test of 25.
[16:45:15.569] [16:45:15] t/027_persistence_change.pl ..........
[16:45:15.569] Dubious, test returned 1 (wstat 256, 0x100)
[16:45:15.569] Failed 1/25 subtests
[16:45:15.569] [16:45:15]
[16:45:15.569]
[16:45:15.569] Test Summary Report
[16:45:15.569] -------------------
[16:45:15.569] t/027_persistence_change.pl (Wstat: 256 Tests: 25 Failed: 1)
[16:45:15.569] Failed test: 18
[16:45:15.569] Non-zero exit status: 1
[16:45:15.569] Files=27, Tests=315, 220 wallclock secs ( 0.14 usr 0.03 sys + 48.94 cusr 17.13 csys = 66.24 CPU)

https://api.cirrus-ci.com/v1/artifact/task/4785083130314752/tap/src/test/recovery/tmp_check/log/regress_log_027_persistence_change

Greetings,

Andres Freund