bench: run rsync over localhost network for fair comparison

Previously rsync ran as a direct local copy (rsync -aH source/ dest/), bypassing the
loopback interface entirely. This meant tc netem latency/loss/rate limits were never
applied to rsync, making the comparison fundamentally unfair.

Now rsync goes through the network via an rsync daemon on localhost:
  rsync --daemon --no-detach --config=rsyncd.conf
  rsync rsync://localhost:PORT/source/ dest/

The daemon is started before rsync tests and killed afterward. Both the rsync client
and FastSync client run under the same systemd-run disk I/O limits when applicable.
This commit is contained in:
2026-07-05 16:58:13 +02:00
parent 7a1c491f7f
commit 5170155112
+96 -55
View File
@@ -6,6 +6,7 @@ import subprocess
import sys
import tempfile
import time
import socket
TEST_DIR = os.path.join(os.path.dirname(os.path.abspath(__file__)), "test_data")
DEFAULT_SOURCE_DIR = os.path.join(TEST_DIR, "source")
@@ -86,6 +87,12 @@ def netem_reset():
)
def find_free_port():
with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
s.bind(('', 0))
return s.getsockname()[1]
def generate_test_files(source_dir):
if os.path.exists(source_dir):
shutil.rmtree(source_dir)
@@ -262,64 +269,98 @@ def run_profile(profile_name, source_dir, dest_dir):
server_process.kill()
server_process.wait()
for case in RSYNC_CASES:
name = case["name"]
rsync_args = case["args"]
print(f"\n --- {name} ---")
# Rsync tests (over network via daemon, so tc netem applies)
rsync_port = find_free_port()
rsyncd_conf = os.path.join(tempfile.gettempdir(), f"rsyncd-{rsync_port}.conf")
with open(rsyncd_conf, "w") as f:
f.write(f"""use chroot = no
max connections = 5
read only = yes
port = {rsync_port}
if os.path.exists(dest_dir):
shutil.rmtree(dest_dir)
[source]
path = {source_dir}
""")
rsync_daemon = None
try:
rsync_daemon = subprocess.Popen(
["rsync", "--daemon", "--no-detach", f"--config={rsyncd_conf}"],
stdout=subprocess.DEVNULL, stderr=None
)
time.sleep(0.5)
for case in RSYNC_CASES:
name = case["name"]
rsync_args = case["args"]
print(f"\n --- {name} ---")
if os.path.exists(dest_dir):
shutil.rmtree(dest_dir)
try:
rsync_cmd = (
client_prefix
+ ["rsync"]
+ rsync_args
+ [f"rsync://localhost:{rsync_port}/source/", f"{dest_dir}/"]
)
print(f" Running: {' '.join(rsync_cmd)}")
start_time = time.monotonic()
rsync_result = subprocess.run(
rsync_cmd, capture_output=True, timeout=120
)
end_time = time.monotonic()
duration = end_time - start_time
mismatches, missing = [], []
if rsync_result.returncode == 0:
mismatches, missing = verify_transfer(source_dir, dest_dir)
entry = {
"name": name,
"suite": profile_name,
"time": f"{duration:.4f}s" if rsync_result.returncode == 0 else "N/A",
}
if rsync_result.returncode == 0 and not mismatches and not missing:
entry["status"] = "Success"
entry["error"] = ""
else:
entry["status"] = "Failed"
errors = []
if rsync_result.returncode != 0:
err = rsync_result.stderr.strip().split("\n")[0] if rsync_result.stderr else "No output"
errors.append(f"Exit code {rsync_result.returncode}: {err[:80]}")
if missing:
errors.append(f"Missing ({len(missing)}): {', '.join(missing[:5])}")
if mismatches:
errors.append(f"Mismatch ({len(mismatches)}): {', '.join(mismatches[:3])}")
entry["error"] = " | ".join(errors)
results.append(entry)
except subprocess.TimeoutExpired:
results.append(
{"name": name, "suite": profile_name, "status": "Timeout", "time": "N/A", "error": "Exceeded 120s"}
)
except Exception as e:
results.append(
{"name": name, "suite": profile_name, "status": "Error", "time": "N/A", "error": str(e)}
)
finally:
if rsync_daemon:
try:
rsync_daemon.wait(timeout=5)
except subprocess.TimeoutExpired:
rsync_daemon.kill()
rsync_daemon.wait()
try:
rsync_cmd = (
["rsync"]
+ rsync_args
+ [f"{source_dir}/", f"{dest_dir}/"]
)
print(f" Running: {' '.join(rsync_cmd)}")
start_time = time.monotonic()
rsync_result = subprocess.run(
rsync_cmd, capture_output=True, timeout=120
)
end_time = time.monotonic()
duration = end_time - start_time
mismatches, missing = [], []
if rsync_result.returncode == 0:
mismatches, missing = verify_transfer(source_dir, dest_dir)
entry = {
"name": name,
"suite": profile_name,
"time": f"{duration:.4f}s" if rsync_result.returncode == 0 else "N/A",
}
if rsync_result.returncode == 0 and not mismatches and not missing:
entry["status"] = "Success"
entry["error"] = ""
else:
entry["status"] = "Failed"
errors = []
if rsync_result.returncode != 0:
err = rsync_result.stderr.strip().split("\n")[0] if rsync_result.stderr else "No output"
errors.append(f"Exit code {rsync_result.returncode}: {err[:80]}")
if missing:
errors.append(f"Missing ({len(missing)}): {', '.join(missing[:5])}")
if mismatches:
errors.append(f"Mismatch ({len(mismatches)}): {', '.join(mismatches[:3])}")
entry["error"] = " | ".join(errors)
results.append(entry)
except subprocess.TimeoutExpired:
results.append(
{"name": name, "suite": profile_name, "status": "Timeout", "time": "N/A", "error": "Exceeded 120s"}
)
except Exception as e:
results.append(
{"name": name, "suite": profile_name, "status": "Error", "time": "N/A", "error": str(e)}
)
os.unlink(rsyncd_conf)
except Exception:
pass
except subprocess.CalledProcessError as e:
print(f" Error running netem command: {' '.join(e.cmd)}")