diff --git a/tests/load/collaboration/run_pgbench.sh b/tests/load/collaboration/run_pgbench.sh
index 07fcd670..05f394cd 100755
--- a/tests/load/collaboration/run_pgbench.sh
+++ b/tests/load/collaboration/run_pgbench.sh
@@ -243,25 +243,30 @@ for raw in "${client_list[@]}"; do
       echo "[pgbench] failed transactions = ${failed:-missing} for $mode/$clients" >&2
       exit 1
     }
-    read -r p50 p95 p99 samples <<<"$(python3 - "$prefix" <<'PY'
-import glob, sys
+    stats="$(python3 - "$prefix" <<'PY'
+import glob, os, re, sys
 prefix = sys.argv[1]
+base = os.path.basename(prefix)
+pat = re.compile(re.escape(base) + r"\.\d+(?:\.\d+)?$")
 vals = []
 for path in sorted(glob.glob(prefix + ".*")):
+    if not pat.match(os.path.basename(path)):
+        continue
     with open(path) as fh:
         for line in fh:
             parts = line.split()
-            if len(parts) >= 3:
+            if len(parts) >= 3 and parts[2].isdigit():
                 vals.append(int(parts[2]))
+if not vals:
+    sys.exit("no latency samples under " + prefix)
 vals.sort()
 def pct(p):
-    if not vals:
-        return "nan"
     idx = int(round((p / 100.0) * (len(vals) - 1)))
     return f"{vals[idx] / 1000.0:.3f}"
 print(pct(50), pct(95), pct(99), len(vals))
 PY
-)"
+)" || exit 1
+    read -r p50 p95 p99 samples <<<"$stats"
     awk -v p50="$p50" 'BEGIN { if (p50+0 > 60000) exit 1 }' || {
       echo "[pgbench] latency sample looks unparsed (p50=${p50}ms)" >&2
       exit 1