Re: Changing the state of data checksums in a running cluster

> On 28 Aug 2026, at 19:00, Alexander Lakhin <exclusion(at)gmail(dot)com> wrote:
>
> Hello Daniel,
>
> 06.04.2026 03:20, Daniel Gustafsson wrote:
>> I've pushed fixes for all of these as well as the intermittent failures that
>> were seen on some BF animals, and will stare at the buildfarm for a while now.
>> So far 10 or so machines have built these green so it looks decent so far.
>
> Yesterday culicidae produced a new failure of the 005_injection.pl test:
> 355/405 test_checksums - postgresql:test_checksums/005_injection FAIL 14.66s exit status 1
>
> [11:07:55.420](0.190s) not ok 3 - ensure checksums are set to off on injection_node
> [11:07:55.421](0.001s) # Failed test 'ensure checksums are set to off on injection_node'
> # at /home/bf/bf-build/culicidae/REL_19_STABLE/pgsql/src/test/modules/test_checksums/t/DataChecksums/Utils.pm line 69.
> [11:07:55.422](0.001s) # got: 'inprogress-off'
> # expected: 'off'

I'll have a go at reproducing the error when I am back in the office, but off
the cuff I would expect the below to fix it. The state transition to off
didn't wait for completion which can lead to that timing error in reading back
the state.

-disable_data_checksums($node);
+disable_data_checksums($node, wait => 1);

> I've also triggered a bit different failure:
> t/005_injection.pl .. 1/?
> # Failed test 'ensure checksums are set to off on injection_node'
> # at .../src/test/modules/test_checksums/t/DataChecksums/Utils.pm line 69.
> # got: 'inprogress-on'
> # expected: 'off'
> t/005_injection.pl .. 15/? # Looks like you failed 1 test of 17.

Interesting, I'll have a closer look when back but test 15 in that file doesn't
test for off at all so the reference to 15 in the above must be a red herring?
Which test was it according to the not ok line in the log?

--
Daniel Gustafsson