Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://github.com/multipath-tcp/mptcp_net-next/issues/336 below:

Issues with BPF packet scheduler · Issue #336 · multipath-tcp/mptcp_net-next · GitHub

Hello,

I'm currently interested in the packet scheduler using BPF and have done some experiments. However, I've run into some issues when using bpf_rr and bpf_red scheduler. There are three possibilities:

1. The client keeps sending data to the server, but the server has received the expected amount of data.
2. The application is stuck, and the ppid becomes 1, which can only be solved by restarting the system. (It seems that the scheduler tries to send data to an unestablished subflow at this time)
3. RST might appear during data transmission.

Problem 1 usually occurs, sometimes problem 2 or 3.

However, using bpf_first scheduler doesn't cause any problems above.

My experiment setup is as follows:

forward@forward-virtual-machine:~$ ip mptcp endpoint show
192.168.2.30 id 1 subflow fullmesh
192.168.2.29 id 2 subflow fullmesh
forward@forward-virtual-machine:~$ ip mptcp limits show
add_addr_accepted 1 subflows 3
forward@forward-virtual-machine:~$ hostnamectl
 Static hostname: forward-virtual-machine
       Icon name: computer-vm
         Chassis: vm
      Machine ID: c74952654b4c49f2b200fed4ca8eb8e1
         Boot ID: 8a19a42e95bf45178099968557309159
  Virtualization: vmware
Operating System: Ubuntu 22.04.1 LTS
          Kernel: Linux 6.1.0+
    Architecture: x86-64
 Hardware Vendor: VMware, Inc.
  Hardware Model: VMware Virtual Platform

forward@forward-virtual-machine:~$ ip mptcp endpoint show
192.168.2.31 id 1 signal
192.168.2.32 id 2 signal
forward@forward-virtual-machine:~$ ip mptcp limits show
add_addr_accepted 1 subflows 3
forward@forward-virtual-machine:~$ hostnamectl
 Static hostname: forward-virtual-machine
       Icon name: computer-vm
         Chassis: vm
      Machine ID: 692e3a02614143b68f61aa642054ddcf
         Boot ID: 570c63dd032c4bd08c42be01322e79fb
  Virtualization: vmware
Operating System: Ubuntu 22.04.1 LTS
          Kernel: Linux 6.1.0+
    Architecture: x86-64
 Hardware Vendor: VMware, Inc.
  Hardware Model: VMware Virtual Platform

We tried to understand this problem more deeply through the BPF tracing tool.

Through stackcount, we found the following question:

sk_reset_timer is being called repeatedly.

We found this question very similar to the one mentioned in #52:

spurious retransmission and the mptcp_worker tries to update the MPTCP retransmission timer

I was wondering if this was the same issue or I made some mistakes?

I'm really open to disclose more information about the issue.
Thank you in advance.

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4