I recently moved from a non docker ubuntu 18.04 setup to a full docker 22.04 setup. I used to get a pretty consistent 95-105 MB/s in sab. I can’t seem to get more than 70 MB/s in my new setup. Both my usenet providers are setup the same and using 50 connections. I’d love to blame docker, but thought i see if anyone here has seen this as well. Any suggestions on what to look at?
One thing I found when testing.my settings was reducing the number of connections can cause better results. Is the speed your getting through the test button or actual downloads? The test also tests other things for the bottleneck so maybe something else is being the bottleneck like disk speeds.
Actual downloads going to the same storage array as the previous system. Same NFS lun even. I’ll try the test button and playing with the number of connections. Thanks! Will report back.
The article your linked is citing a paper from 2014. Seems hard the believe the network stack hasn’t been significantly reworked in 10 years. That being said I don’t have anything to refute this.
Does the above still apply if using a glutun VPN docker with other containers connecting directly to glutun?
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !selfhosted@lemmy.world
A place to share alternatives to popular online services that can be self-hosted without giving up privacy or locking you into a service you don’t control.
Rules:
Be civil: we’re here to support and learn from one another. Insults won’t be tolerated. Flame wars are frowned upon.
No spam posting.
Posts have to be centered around self-hosting. There are other communities for discussing hardware or home computing. If it’s not obvious why your post topic revolves around selfhosting, please include details to make it clear.
Don’t duplicate the full text of your blog or github here. Just post the link for folks to click.
Submission headline should match the article title (don’t cherry-pick information from the title to fit your agenda).
No trolling.
Resources:
selfh.st Newsletter and index of selfhosted software and apps
One thing I found when testing.my settings was reducing the number of connections can cause better results. Is the speed your getting through the test button or actual downloads? The test also tests other things for the bottleneck so maybe something else is being the bottleneck like disk speeds.
Actual downloads going to the same storage array as the previous system. Same NFS lun even. I’ll try the test button and playing with the number of connections. Thanks! Will report back.
deleted by creator
This could be the I/O overhead caused by docker while using NAT networking: https://stackoverflow.com/questions/21889053/what-is-the-runtime-performance-cost-of-a-docker-container#26149994
The article your linked is citing a paper from 2014. Seems hard the believe the network stack hasn’t been significantly reworked in 10 years. That being said I don’t have anything to refute this.
Does the above still apply if using a glutun VPN docker with other containers connecting directly to glutun?
That’s a large penalty, wow.
It’s huge for networking. VPN gateway (for accessing the VPN WAN over LAN) on the host is 20-40% faster than dockerized…
I’m using NAT for sure. All my containers are on a different subnet than the host.