roachtest: unoptimized-query-oracle/disable-rules=all/seed-multi-region failed

[cockroach-teamcity]

roachtest.unoptimized-query-oracle/disable-rules=all/seed-multi-region failed with artifacts on release-23.1.9-rc.FROZEN @ 28b2e767986169df857f5f2cd1790d71ac633395:

  	strings.Join({
  		"704473406,-1201.3111417457",
- 		"808",
+ 		"926",
  		",false",
  	}, ""),
  	strings.Join({
  		"704473406,-122.4493691481332",
- 		"6",
  		",true",
  	}, ""),
  	strings.Join({
  		"704473406,-123.3310999792810",
- 		"2",
+ 		"3",
  		",NULL",
  	}, ""),
  	strings.Join({
  		"704473406,-1231.848445144695",
- 		"2",
+ 		"7",
  		",true",
  	}, ""),
  	strings.Join({
  		"704473406,-1299.5378016985",
- 		"55",
+ 		"60",
  		"5,false",
  	}, ""),
  	strings.Join({
  		"704473406,-13.6629269156472",
- 		"47",
+ 		"19",
  		",true",
  	}, ""),
  	... // 696 identical, 19 removed, 19 inserted, and 542 modified elements
  }
sql: SELECT
	" tab67"._int4 AS col256,
	sum("t\\x8fab/65"._float8::FLOAT8)::FLOAT8 AS "co\\xf0l257",
	bool_or("ta-b66"._bool::BOOL)::BOOL AS "c�ol258"
FROM
	defaultdb.public.seed_mr_table@[0] AS "t\\x8fab/65"
	JOIN defaultdb.public.seed_mr_table@[0] AS "ta-b66" ON ("t\\x8fab/65"._date) = ("ta-b66"._date)
	JOIN defaultdb.public.seed_mr_table@[0] AS " tab67" ON ("ta-b66"._int8) = (" tab67"._int2)
GROUP BY
	"ta-b66"._bool, "t\\x8fab/65"._float8, " tab67"._int4
ORDER BY
	" tab67"._int4 ASC NULLS LAST
test artifacts and logs in: /artifacts/unoptimized-query-oracle/disable-rules=all/seed-multi-region/run_1

Parameters: ROACHTEST_arch=amd64 , ROACHTEST_cloud=gce , ROACHTEST_cpu=4 , ROACHTEST_encrypted=false , ROACHTEST_ssd=0

Help

See: roachtest README

See: How To Investigate (internal)

See: Grafana

/cc @cockroachdb/sql-queries

This test on roachdash | Improve this report!

Jira issue: CRDB-31298

[rharding6373]

These differences are at the 15th significant digit or higher for an aggregation on floats. We could try to backport #106552, which would have correctly noted that these are approximately equal values or continue to manually identify approximate floats. Either way, not a release blocker.

[rharding6373]

I opened some backports. Closing since this is not a real error.