fix: Move bench bucket to eu (#2615)

Author: robert3005

.github/workflows/generate-tpch-s3.yml

@@ -20,7 +20,7 @@ jobs:
       - self-hosted
       - aws
     env:
-      REMOTE_PATH: s3://vortex-bench-dev/tpch-sf1/
+      REMOTE_PATH: s3://vortex-bench-dev-eu/tpch-sf1/
       TMPDIR: /work
     steps:
       - uses: actions/checkout@v4

.github/workflows/sql-benchmarks.yml

@@ -24,7 +24,7 @@ jobs:
           - id: tpch-s3
             binary_name: tpch
             name: TPC-H on S3
-            remote_storage: s3://vortex-bench-dev/tpch-sf1/
+            remote_storage: s3://vortex-bench-dev-eu/tpch-sf1/
     runs-on:
       - runs-on=${{ github.run_id }}
       - family=c7i.8xlarge
@@ -61,6 +61,7 @@ jobs:
         shell: bash
         env:
           RUSTFLAGS: '-C target-cpu=native'
+          AWS_REGION: 'eu-west-1'
           OTEL_SERVICE_NAME: 'vortex-bench'
           OTEL_EXPORTER_OTLP_PROTOCOL: 'http/protobuf'
           OTEL_EXPORTER_OTLP_ENDPOINT: '${{ secrets.OTEL_EXPORTER_OTLP_ENDPOINT }}'

bench-vortex/src/bin/tpch.rs

@@ -135,14 +135,14 @@ fn main() -> ExitCode {
             .unwrap()
         }
         Some(tpch_benchmark_remote_data_dir) => {
-            // e.g. "s3://vortex-bench-dev/parquet/"
+            // e.g. "s3://vortex-bench-dev-eu/parquet/"
             //
             // The trailing slash is significant!
             //
             // The folder must already be populated with data!
             if !tpch_benchmark_remote_data_dir.ends_with("/") {
                 warn!(
-                    "Supply a --use-remote-data-dir argument which ends in a slash e.g. s3://vortex-bench-dev/parquet/"
+                    "Supply a --use-remote-data-dir argument which ends in a slash e.g. s3://vortex-bench-dev-eu/parquet/"
                 );
             }
             info!(

bench-vortex/src/tpch/mod.rs

@@ -12,6 +12,7 @@ use datafusion::datasource::listing::{
 };
 use datafusion::prelude::{CsvReadOptions, ParquetReadOptions, SessionContext};
 use futures::StreamExt;
+use log::info;
 use named_locks::with_lock;
 use object_store::ObjectStore;
 use object_store::aws::AmazonS3Builder;
@@ -236,11 +237,11 @@ async fn register_parquet(
     let mut pq_file = file.clone();
     pq_file.set_path(&pq_file.path().to_owned().replace(".tbl", ".parquet"));
 
-    if object_store
+    if let Err(e) = object_store
         .head(&ObjectStorePath::parse(pq_file.path())?)
         .await
-        .is_err()
     {
+        info!("File {} doesn't exist because {e}", pq_file);
         with_lock(pq_file.path().to_owned(), async || {
             let df = session
                 .read_csv(
@@ -304,11 +305,11 @@ async fn register_vortex_file(
 
     let vtx_file = &vortex_dir.join(vortex_basename.as_ref())?;
 
-    if object_store
+    if let Err(e) = object_store
         .head(&ObjectStorePath::parse(vtx_file.path())?)
         .await
-        .is_err()
     {
+        info!("File {} doesn't exist because {e}", vtx_file);
         with_lock(vtx_file.path().to_owned(), async || {
             let record_batches = session
                 .read_csv(

deny.toml

@@ -11,8 +11,7 @@ feature-depth = 1
 
 
 [advisories]
-# ring is unmaintained: https://github.com/briansmith/ring/discussions/2414
-ignore = ["RUSTSEC-2025-0007"]
+ignore = ["RUSTSEC-2025-0009"]
 
 [licenses]
 # List of explicitly allowed licenses