cronekorkn/left4me
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

docs(deploy): tighten perf-tuning escape hatches

Browse source 
- RT example: add AmbientCapabilities=CAP_SYS_NICE so the User=left4me
  service can actually enter SCHED_FIFO on Trixie.
- CPU governor: note that linux-cpupower may need apt install.
- CPUAffinity=2: clarify that per-instance values typically increment.
- NIC tuning: note that ethtool may need apt install.
This commit is contained in:
mwiegand 2026-05-09 10:15:45 +02:00
parent 9e0f6f17ef
commit e5126c8c0b
No known key found for this signature in database
1 changed files with 8 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
deploy/README.md
View file

@ -86,6 +86,8 @@ The performance governor squeezes a few percent off jitter under bursty load. `s
sudo cpupower frequency-set -g performance sudo cpupower frequency-set -g performance
``` ```
Install via `sudo apt install linux-cpupower` if the binary isn't present.
Persist via your distro's CPU-frequency tooling (e.g. `/etc/default/cpufrequtils`). Persist via your distro's CPU-frequency tooling (e.g. `/etc/default/cpufrequtils`).
### Per-instance CPU affinity ### Per-instance CPU affinity
@ -97,11 +99,13 @@ Persist via your distro's CPU-frequency tooling (e.g. `/etc/default/cpufrequtils
CPUAffinity=2 CPUAffinity=2
``` ```
This pins the instance to CPU 2 specifically; per-instance values would typically be 1, 2, 3, ... so each server has its own core.
A reasonable strategy on an N-core host: leave core 0 for the kernel + IRQs + system services, then pin one instance per remaining core. A reasonable strategy on an N-core host: leave core 0 for the kernel + IRQs + system services, then pin one instance per remaining core.
### NIC tuning ### NIC tuning
Hardware-specific. On a host with a single primary interface (replace `eth0`): Hardware-specific (install via `sudo apt install ethtool` if not present). On a host with a single primary interface (replace `eth0`):
```sh ```sh
sudo ethtool -G eth0 rx 4096 tx 4096 sudo ethtool -G eth0 rx 4096 tx 4096
@ -121,8 +125,11 @@ Source-engine servers do not need real-time scheduling, and a misbehaving `srcds
CPUSchedulingPolicy=fifo CPUSchedulingPolicy=fifo
CPUSchedulingPriority=10 CPUSchedulingPriority=10
LimitRTPRIO=10 LimitRTPRIO=10
AmbientCapabilities=CAP_SYS_NICE
``` ```
The `AmbientCapabilities=CAP_SYS_NICE` line is needed because the service runs as `User=left4me` with `NoNewPrivileges=true`; without it some kernels/systemd combinations refuse to apply the RT policy.
### Applying changes to running servers ### Applying changes to running servers
Unit-file changes do not apply to already-running services. After any change: Unit-file changes do not apply to already-running services. After any change: