I don't completely understand it myself. I have a Focal system with k3s installed and about 70 pods (about 150 containers) and about 70 veth devices.
In /lib/systemd/system/systemd-networkd.socket under "[Socket]" I have set "ReceiveBuffer=128M". Maybe that "128M" isn't enough but I can't add any more pods without getting "Could not enumerate addresses: No buffer space available" errors, so I *assumed* there's more to that patch than just the ReceiveBuffer setting. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1912331 Title: Many interfaces lead to "kernel receive buffer overrun" To manage notifications about this bug go to: https://bugs.launchpad.net/systemd/+bug/1912331/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
