From fc66c254906d2eaadcdae60dd4927e1e3af80289 Mon Sep 17 00:00:00 2001 From: AztecBot Date: Thu, 11 Jun 2026 18:41:26 +0000 Subject: [PATCH] fix(ci): bump avm_check_circuit per-test timeout to 120s for heavy txs --- yarn-project/end-to-end/bootstrap.sh | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/yarn-project/end-to-end/bootstrap.sh b/yarn-project/end-to-end/bootstrap.sh index 011679338e8f..205cb810c70e 100755 --- a/yarn-project/end-to-end/bootstrap.sh +++ b/yarn-project/end-to-end/bootstrap.sh @@ -198,11 +198,14 @@ function avm_check_circuit_cmds { # Commands run from repo root via parallelize, so use path from top local dump_dir_from_top="yarn-project/end-to-end/$default_avm_inputs_dump_dir" - # Specify timeout and resources - # WARNING: theoretically, transactions could need more CPU and MEM than we allocate by default. - # In that case, they might start timing out. For now, all of the e2e test txs seem to be relatively - # small and the AVM can run check-circuit with limited resources. - local prefix="$hash:ISOLATE=1:TIMEOUT=30s" + # Specify timeout and resources. + # Most e2e txs are small and check-circuit finishes in a few seconds, but some heavier txs + # (e.g. one of the e2e_multiple_blobs txs) generate ~700k-row traces whose simulation, trace + # generation, and circuit check take well over 30s on the default 2 CPUs and so time out at the + # boundary. Use a generous timeout so the heaviest txs have headroom. This is zero-cost for the + # fast majority: they still exit in a few seconds, so parallelism and stage wall-clock are + # unaffected -- only the heavy outlier is allowed to run to completion instead of being killed. + local prefix="$hash:ISOLATE=1:TIMEOUT=120s" # Find all .bin files in the dump directory (handles nested dirs) for input_file in "$default_avm_inputs_dump_dir"/*/*.bin "$default_avm_inputs_dump_dir"/*/*/*.bin; do