Proxmox, SMB LXC, Debian VM + Docker + Traefik

As I self-host some services, for fun and testing, I switched from NPM to Traefik recently.
Configuration is easier, tutorials are easy to follow (for the basis) and community seems really reactive.

I noticed, as many peoples, that Traefik slow backend response time and reduce throughput...

I managed to use Varnish as caching backend (point loadBalancer to Varnish instance) as in my vcl I tell Varnish to backend depending on hostname...

Well it's like this:
Internet -> Cloudflare -> [home] -> Proxmox -> VM -> Docker -> Traefik -> Varnish -> Real backend

It works good.

But I noticed a little optimisation on tweaking sysctl and limits..

sysctl.conf

fs.file-max=9999999
fs.nr_open=9999999
net.core.netdev_max_backlog=4096
net.core.rmem_max=16777216
net.core.somaxconn=65535
net.core.wmem_max=16777216
net.ipv4.ip_local_port_range=1025 65535
net.ipv4.tcp_fin_timeout=30
net.ipv4.tcp_keepalive_time=30
net.ipv4.tcp_max_syn_backlog=20480
net.ipv4.tcp_max_tw_buckets=400000
net.ipv4.tcp_no_metrics_save=1
net.ipv4.tcp_syn_retries=2
net.ipv4.tcp_synack_retries=2
net.ipv4.tcp_tw_recycle=1
net.ipv4.tcp_tw_reuse=1
vm.min_free_kbytes=65536
vm.overcommit_memory=1

limits.conf

  • soft nofile 9999999
  • hard nofile 9999999
    root soft nofile 9999999
    root hard nofile 9999999

Maybe some insecure setting this so high, but response time seems better.

If it can help :slight_smile:

1 Like