berrie64 always failing on interrupt tests

[robe]

berrie64 has been failing consistently on interrupt tests for past couple of runs ​https://debbie.postgis.net/view/PostGIS/job/PostGIS_Worker_Run/5759/label=berrie64/consoleFull

00:01:44 PostgreSQL 17devel on aarch64-unknown-linux-gnu, compiled by gcc (Debian 10.2.1-6) 10.2.1 20210110, 64-bit
00:01:44   Postgis 3.5.0dev - (4057a91) - 2024-03-11 02:12:26
00:01:44   scripts 3.5.0dev 4057a91
00:01:44   GEOS: 3.13.0dev-CAPI-1.18.0
00:01:44   PROJ: 7.2.1 NETWORK_ENABLED=OFF URL_ENDPOINT=https://cdn.proj.org USER_WRITABLE_DIRECTORY=/home/jenkins/.local/share/proj DATABASE_PATH=/usr/share/proj/proj.db
00:01:44   Most PostGIS functions locked into view usage

00:05:47  regress/core/interrupt_relate .. failed (diff expected obtained: /tmp/pgis_reg/test_123_diff)
00:06:16 -----------------------------------------------------------------------------
00:06:16 --- ./core/interrupt_relate_expected	2024-03-10 20:11:13.578793975 -0600
00:06:16 +++ /tmp/pgis_reg/test_123_out	2024-03-10 22:06:16.037211437 -0600
00:06:16 @@ -1,5 +1,5 @@
00:06:16  ERROR:  canceling statement due to statement timeout
00:06:16 -contains interrupted on time
00:06:16 +contains interrupted late: 00:00:14.510019 (00:00:02.01476 tolerated)
00:06:16  ERROR:  canceling statement due to statement timeout
00:06:16  covers interrupted on time
00:06:16  ERROR:  canceling statement due to statement timeout
00:06:16 -----------------------------------------------------------------------------

Not sure if this signals some sort of issue with PG17 or if I should just turn the interrupt tests off, as they are usually false negatives.

[pramsey]

Still?

[robe]

I think I might have turned it off haven't seen interrupt tests recently failing.