From 6a4a6476791f484ff57fcfc1ef84d56b4709ecfe Mon Sep 17 00:00:00 2001 From: shuke987 Date: Mon, 15 Jun 2026 17:56:58 +0800 Subject: [PATCH] [fix](regression) keep variant_doc_mode_p2 load in the p2 group with its queries `variant_doc_mode_p2/load.groovy` (which creates and loads the `github_events` table) was tagged `nonConcurrent,p2`, while the 76 `sql/*.sql` query cases under the same directory fall back to the path-derived default group `variant_doc_mode_p2` (no `nonConcurrent`). Because `nonConcurrent` is a separate pipeline, the load and the queries get routed to different runs/clusters. In the P2 run the load step never executes, so every query fails instantly with `Table [github_events] does not exist in database [regression_test_variant_doc_mode_p2]` (66 cascade failures in one build). The load only sets session variables and does no global state mutation (`set global` / FE/BE config / debug points), so it does not need `nonConcurrent` isolation. Drop the tag so load and queries share the `p2` group; `load.groovy` sorts before `sql/`, runs first, and the table is present when the queries run. Co-Authored-By: Claude Opus 4.8 (1M context) --- regression-test/suites/variant_doc_mode_p2/load.groovy | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/regression-test/suites/variant_doc_mode_p2/load.groovy b/regression-test/suites/variant_doc_mode_p2/load.groovy index e9a926f380e12d..03bd0e3f4cfc6a 100644 --- a/regression-test/suites/variant_doc_mode_p2/load.groovy +++ b/regression-test/suites/variant_doc_mode_p2/load.groovy @@ -17,7 +17,7 @@ import org.codehaus.groovy.runtime.IOGroovyMethods -suite("test_doc_value_p2", "nonConcurrent,p2"){ +suite("test_doc_value_p2", "p2"){ sql """ set default_variant_enable_doc_mode = true """ boolean random = new Random().nextBoolean() if (random) {