Netatalk 4.4.0
Name
afp_lantest — AFP LAN performance and directory cache testing tool
Synopsis
afp_lantest [-34567bcGgKVv] [-h host] [-p port] [-s volume] [-u user] [-w password] [-n iterations] [-f tests] [-F bigfile]
Description
afp_lantest is a comprehensive AFP (Apple Filing Protocol) performance testing tool designed to benchmark various aspects of AFP servers, including directory cache performance in netatalk.
The tool runs a series of tests that measure file operations, directory traversal, and caching efficiency. It includes both traditional file system benchmarks and specialized cache-focused tests that highlight the benefits of optimized directory cache validation and probabilistic validation features.
Options
-3 : Use AFP 3.0 protocol version
-4 : Use AFP 3.1 protocol version
-5 : Use AFP 3.2 protocol version
-6 : Use AFP 3.3 protocol version
-7 : Use AFP 3.4 protocol version (default)
-b : Debug mode
-c : Output results in CSV format (default: tabular)
-F bigfile : Use existing file in volume root for read test (file size must match -g/-G options)
-f tests : Specific tests to run, specified as digits (e.g., “134” runs tests 1, 3, and 4)
-G : Optimize for 10-gigabit networks (increases file test size to 10 GB)
-g : Optimize for gigabit networks (increases file test size to 1 GB)
-h host : Server hostname or IP address (default: localhost)
-K : Run cache-focused tests only (tests 9-12)
-n iterations : Number of test iterations to run (default: 2). For iterations > 5 outliers will be removed
-p port : Server port number (default: 548)
-s volume : Volume name to mount for testing
-u user : Username for authentication with AFP server (default: current uid)
-V : Very verbose output
-v : Verbose output
-w password : Password for authentication with AFP server
Configuration
The test runner AFP client only supports the ClearTxt UAM currently. Configure the UAM in netatalk’s afp.conf:
[Global]
uam list = uams_clrtxt.so
Available Tests
The following tests are available:
(1) Opening, stating and reading 512 bytes from 1000 files : Tests basic file access patterns with many small operations
(2) Writing one large file : Measures sustained write performance
(3) Reading one large file : Measures sustained read performance
(4) Locking/Unlocking 10000 times each : Tests file locking performance
(5) Creating dir with 2000 files : Tests directory creation and file creation performance
(6) Enumerate dir with 2000 files : Tests directory enumeration with many files
(7) Deleting dir with 2000 files : Tests directory and file deletion performance
(8) Create directory tree with 1000 dirs : Tests nested directory creation
(9) Directory cache hits [CACHE] : Tests directory and file lookup performance (1000 dirs + 10000 files)
(10) Mixed cache operations [CACHE] : Tests mixed create/stat/enum/delete operations
(11) Deep path traversal [CACHE] : Tests performance with deep directory structures
(12) Cache validation [CACHE] : Tests directory cache validation mechanisms
Cache-Focused Tests
Tests 9-12 are specifically designed to highlight directory cache performance improvements in netatalk. These tests benefit significantly from optimized cache validation and probabilistic validation features. Use the -K option to run only these cache-focused tests.
Examples
Run all tests with default settings:
afp_lantest -h server.example.com -u testuser -w password -s TestVolume
Run only cache-focused tests:
afp_lantest -K -h server.example.com -u testuser -w password -s TestVolume
Run specific tests (file operations):
afp_lantest -f 123 -h server.example.com -u testuser -w password -s TestVolume
Run tests optimized for gigabit network:
afp_lantest -g -h server.example.com -u testuser -w password -s TestVolume
Run multiple iterations for statistical analysis:
afp_lantest -n 5 -h server.example.com -u testuser -w password -s TestVolume
Example with IO Monitoring
IO Monitoring (Linux only) requires proc filesystem mounted at /proc_io with hidepid=0. Set gid to the group ID of the user running afp_lantest.
For example, to run as root;
mkdir -p /proc_io && mount -t proc -o hidepid=0,gid=0 proc /proc_io
afp_lantest -n 2 -7 -h 127.0.0.1 -p 548 -u test -w test -s 'File Sharing'
Connecting to host 127.0.0.1:548
IO monitoring: /proc_io is available
Found cnid_dbd process for user 'test': PID 40
Found privilege-dropped afpd process for user 'test': PID 36
IO monitoring enabled (afpd: 36, cnid_dbd: 40)
Run 1 => Open, stat and read 512 bytes from 1000 files [8,000 AFP ops] 4230 ms
IO Operations; afpd: 6000 READs, 7002 WRITEs | cnid_dbd: 0 READs, 2 WRITEs
Run 1 => Writing one large file [103 AFP ops] 166 ms for 100 MB (avg. 631 MB/s)
IO Operations; afpd: 0 READs, 299 WRITEs | cnid_dbd: 0 READs, 0 WRITEs
Run 1 => Reading one large file [102 AFP ops] 55 ms for 100 MB (avg. 1906 MB/s)
IO Operations; afpd: 100 READs, 100 WRITEs | cnid_dbd: 0 READs, 0 WRITEs
Run 1 => Locking/Unlocking 10000 times each [20,000 AFP ops] 859 ms
IO Operations; afpd: 0 READs, 20000 WRITEs | cnid_dbd: 0 READs, 0 WRITEs
Run 1 => Creating dir with 2000 files [4,000 AFP ops] 4927 ms
IO Operations; afpd: 2000 READs, 10006 WRITEs | cnid_dbd: 4 READs, 6151 WRITEs
Run 1 => Enumerate dir with 2000 files [~51 AFP ops] 1740 ms
IO Operations; afpd: 1959 READs, 50 WRITEs | cnid_dbd: 0 READs, 2 WRITEs
Run 1 => Deleting dir with 2000 files [2,000 AFP ops] 3362 ms
IO Operations; afpd: 4000 READs, 4004 WRITEs | cnid_dbd: 4 READs, 6177 WRITEs
Run 1 => Create directory tree with 1000 dirs [1,110 AFP ops] 2081 ms
IO Operations; afpd: 0 READs, 4445 WRITEs | cnid_dbd: 2 READs, 2279 WRITEs
Run 1 => Directory cache hits (100 dirs + 1000 files) [11,000 AFP ops] 5552 ms
IO Operations; afpd: 10000 READs, 11100 WRITEs | cnid_dbd: 0 READs, 100 WRITEs
Run 1 => Mixed cache operations (create/stat/enum/delete) [820 AFP ops] 1215 ms
IO Operations; afpd: 820 READs, 1623 WRITEs | cnid_dbd: 0 READs, 1203 WRITEs
Run 1 => Deep path traversal (nested directory navigation) [3,500 AFP ops] 1835 ms
IO Operations; afpd: 2500 READs, 3550 WRITEs | cnid_dbd: 0 READs, 50 WRITEs
Run 1 => Cache validation efficiency (metadata changes) [30,000 AFP ops] 14559 ms
IO Operations; afpd: 30000 READs, 30100 WRITEs | cnid_dbd: 0 READs, 100 WRITEs
Run 2 => Open, stat and read 512 bytes from 1000 files [8,000 AFP ops] 3801 ms
IO Operations; afpd: 6000 READs, 7005 WRITEs | cnid_dbd: 0 READs, 5 WRITEs
Run 2 => Writing one large file [103 AFP ops] 115 ms for 100 MB (avg. 911 MB/s)
IO Operations; afpd: 0 READs, 299 WRITEs | cnid_dbd: 0 READs, 0 WRITEs
Run 2 => Reading one large file [102 AFP ops] 47 ms for 100 MB (avg. 2231 MB/s)
IO Operations; afpd: 100 READs, 100 WRITEs | cnid_dbd: 0 READs, 0 WRITEs
Run 2 => Locking/Unlocking 10000 times each [20,000 AFP ops] 1061 ms
IO Operations; afpd: 0 READs, 20000 WRITEs | cnid_dbd: 0 READs, 0 WRITEs
Run 2 => Creating dir with 2000 files [4,000 AFP ops] 4739 ms
IO Operations; afpd: 2000 READs, 10005 WRITEs | cnid_dbd: 7 READs, 6141 WRITEs
Run 2 => Enumerate dir with 2000 files [~51 AFP ops] 1279 ms
IO Operations; afpd: 1959 READs, 50 WRITEs | cnid_dbd: 0 READs, 1 WRITEs
Run 2 => Deleting dir with 2000 files [2,000 AFP ops] 3470 ms
IO Operations; afpd: 4000 READs, 4004 WRITEs | cnid_dbd: 4 READs, 6182 WRITEs
Run 2 => Create directory tree with 1000 dirs [1,110 AFP ops] 2030 ms
IO Operations; afpd: 0 READs, 4445 WRITEs | cnid_dbd: 2 READs, 2272 WRITEs
Run 2 => Directory cache hits (100 dirs + 1000 files) [11,000 AFP ops] 7074 ms
IO Operations; afpd: 10000 READs, 11100 WRITEs | cnid_dbd: 0 READs, 100 WRITEs
Run 2 => Mixed cache operations (create/stat/enum/delete) [820 AFP ops] 1227 ms
IO Operations; afpd: 820 READs, 1622 WRITEs | cnid_dbd: 2 READs, 1243 WRITEs
Run 2 => Deep path traversal (nested directory navigation) [3,500 AFP ops] 1258 ms
IO Operations; afpd: 2500 READs, 3550 WRITEs | cnid_dbd: 0 READs, 50 WRITEs
Run 2 => Cache validation efficiency (metadata changes) [30,000 AFP ops] 20181 ms
IO Operations; afpd: 30000 READs, 30100 WRITEs | cnid_dbd: 0 READs, 100 WRITEs
Successfully deleted test directory 'LanTest-35'
Netatalk Lantest Results (Averages and standard deviations (±) for all tests, across 2 iterations (default))
============================================================================================================
Test Time_ms Time± AFPD_R AFPD_R± AFPD_W AFPD_W± CNID_R CNID_R± CNID_W CNID_W± MB/s
------------------------------------------------------------------ -------- ------ ------ ------- ------ ------- ------ ------- ------ ------- ------
Open, stat and read 512 bytes from 1000 files [8,000 AFP ops] 4015 303.3 6000 0.0 7003 2.2 0 0.0 3 2.2 0
Writing one large file [103 AFP ops] 140 36.1 0 0.0 299 0.0 0 0.0 0 0.0 714
Reading one large file [102 AFP ops] 51 5.7 100 0.0 100 0.0 0 0.0 0 0.0 1960
Locking/Unlocking 10000 times each [20,000 AFP ops] 960 142.8 0 0.0 20000 0.0 0 0.0 0 0.0 0
Creating dir with 2000 files [4,000 AFP ops] 4833 132.9 2000 0.0 10005 1.0 5 2.2 6146 7.1 0
Enumerate dir with 2000 files [~51 AFP ops] 1509 326.0 1959 0.0 50 0.0 0 0.0 1 1.0 0
Deleting dir with 2000 files [2,000 AFP ops] 3416 76.4 4000 0.0 4004 0.0 4 0.0 6179 3.6 0
Create directory tree with 1000 dirs [1,110 AFP ops] 2055 36.1 0 0.0 4445 0.0 2 0.0 2275 5.0 0
Directory cache hits (100 dirs + 1000 files) [11,000 AFP ops] 6313 1076.2 10000 0.0 11100 0.0 0 0.0 100 0.0 0
Mixed cache operations (create/stat/enum/delete) [820 AFP ops] 1221 8.5 820 0.0 1622 1.0 2 0.0 1223 28.3 0
Deep path traversal (nested directory navigation) [3,500 AFP ops] 1546 408.0 2500 0.0 3550 0.0 0 0.0 50 0.0 0
Cache validation efficiency (metadata changes) [30,000 AFP ops] 17370 3975.4 30000 0.0 30100 0.0 0 0.0 100 0.0 0
------------------------------------------------------------------ -------- ------ ------ ------- ------ ------- ------ ------- ------ ------- ------
Sum of all AFP OPs = 80686 43429 57379 92278 13 16077
Aggregates Summary:
------------------------------------------------------------------
Average Time per AFP OP: 0.538 ms
Average AFPD Reads per AFP OP: 0.711
Average AFPD Writes per AFP OP: 1.144
See afp_lantest manpage for more information: https://netatalk.io/manual/en/afp_lantest.1
Dircache Statistics (/var/log/afpd.log):
------------------------------------------------------------------
Sep 24 13:30:15.702673 afpd[36] {dircache.c:632} (info:AFPDaemon): dircache statistics: (user: test) entries: 0, lookups: 244476, hits: 227977 (93.3%), misses: 9228, added: 9552, removed: 9552, expunged: 7271, evicted: 0
IO Monitoring Result Columns
Time(ms) = Test runtime in milliseconds
Time± = Test runtime standard deviation
AFPD_R = afpd process IO Read operations
AFPD_R± = afpd process IO Read operation standard deviation
AFPD_W = afpd process IO Write operations
AFPD_W± = afpd process IO Write operation standard deviation
CNID_* = IO measurements for the cnid_dbd process (optional)
Note; When measuring afpd read/write IO with afp_lantest, ensure afp.conf log level is set to default:error
- More verbose logging superficially increases *_W IO values. For dirstats at least
default:infois reqired.
IO Monitoring Aggregates Summary
The aggregate values are purely Intrinsic Metrics, as AFP operations are a mixture of reads, writes, and connection related operations.
Therefore the measurements are meaningful only within their own context. They provide a way to assess changes relative to itself, but lack any external coherence or reference point for comparison with other systems.
Generally, values less than 1 indicate efficient operation (for example due to batching and caching etc), and values greater than 1 indicate sub-optimal operation (for example amplification of a single operation, causing additional downstream operations).
Note; As total AFP Ops are a mix of reads, writes, and other ops the effective 1:1 (AFP:IO) thresholds are not equal to 1. Accurate thresholds can be calculated on a per test basis using the information shown in the Testing Wiki, and comparing relevant AFP read/write ops with the relevant AFPD_R/AFPD_W values.
Ideally code changes should observe aggregate values as reducing.
For more information see Testing Wiki
Return Codes
0 : All tests passed successfully
1 : One or more tests failed
Notes
- Tests that involve large file operations (tests 2 and 3) will create temporary files in the test volume
- The -g and -G options significantly increase test file sizes and test duration
- Cache-focused tests (8-11) provide the most insight into netatalk’s directory cache performance
- Multiple iterations (-n) are recommended for consistent performance measurements
- The tool requires write access to the specified AFP volume
See Also
afp_speedtest(1), afpd(8), netatalk(8)
