[FS#4161] RTL838x: Netgear GS108T v3 - Ring contention

Sun Nov 28 00:12:51 PST 2021

THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.

A new Flyspray task has been opened.  Details are below. 

User who did this - Torxgewinde (Torxgewinde) 

Attached to Project - OpenWrt/LEDE Project
Summary - RTL838x: Netgear GS108T v3 - Ring contention
Task Type - Bug Report
Category - Base system
Status - Unconfirmed
Assigned To - 
Operating System - All
Severity - Low
Priority - Very Low
Reported Version - openwrt-21.02
Due in Version - Undecided
Due Date - Undecided
Details - ==Device problem occurs on:==
Netgear GS108T v3

==Software versions of OpenWrt/LEDE release, packages, etc.:==
Release 21.01.1 + packages: collectd-mod-ping luci-app-statistics collectd-mod-uptime htop iperf3 

==Steps to reproduce:==
Also posted in the OpenWRT forum: [[https://forum.openwrt.org/t/rtl838x-netgear-gs108t-v3-ring-contention/111421]].

A switch (Netgear GS108T v3, 21.01.1) shows warnings in the log and drops packets aimed towards the device. Traffic entering a port and leaving another port is unaffected/fine, that is why i consider this a low prio issue, but a potential issue nevertheless:

Tue Nov  9 17:29:43 2021 kern.warn kernel: [88189.851532] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a700003c, cur a700004c
Tue Nov  9 17:29:43 2021 kern.warn kernel: [88189.866290] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a7000040, cur a700004c
Tue Nov  9 17:29:43 2021 kern.warn kernel: [88189.889970] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a7000044, cur a700004c
Tue Nov  9 17:29:43 2021 kern.warn kernel: [88189.969803] rtl838x-eth bb00a300.ethernet eth0: Ring 
...
Tue Nov  9 17:30:28 2021 kern.info kernel: [88234.247018] RX buffer overrun: status 101, mask: ffeff
Tue Nov  9 17:30:28 2021 kern.info kernel: [88234.252886] RX buffer overrun: status 1, mask: ffeff
Tue Nov  9 17:30:28 2021 kern.info kernel: [88234.258510] RX buffer overrun: status 1, mask: ffeff

The error just occurs after a while: [[https://forum.openwrt.org/uploads/default/optimized/3X/7/0/70e3354db9cc8eae74256ed160e659ef124bc072_2_1035x210.png|Graph of Ping Log]]

The issue can be triggered by running iperf3:

iperf3 -c 192.168.xx.254 -t0
Connecting to host 192.168.xx.254, port 5201
[  5] local 192.168.xx.13 port 44924 connected to 192.168.xx.254 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec  4.69 MBytes  39.2 Mbits/sec    0    233 KBytes       
[  5]   1.00-2.05   sec  2.75 MBytes  22.1 Mbits/sec    0    369 KBytes       
[  5]   2.05-3.26   sec  3.57 MBytes  24.7 Mbits/sec    0    533 KBytes       
[  5]   3.26-4.03   sec  1.12 MBytes  12.3 Mbits/sec    1    551 KBytes       
[  5]   4.03-5.00   sec  0.00 Bytes  0.00 bits/sec    1    551 KBytes       
[  5]   5.00-6.00   sec  1.11 MBytes  9.34 Mbits/sec    0    420 KBytes       
[  5]   6.00-7.00   sec  1.37 MBytes  11.5 Mbits/sec    0    451 KBytes       
...
[  5]  99.00-100.00 sec  0.00 Bytes  0.00 bits/sec    0   1.72 MBytes       
[  5] 100.00-101.00 sec  0.00 Bytes  0.00 bits/sec    0   1.72 MBytes       
^C[  5] 101.00-101.93 sec  0.00 Bytes  0.00 bits/sec    0   1.72 MBytes       
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-101.93 sec  66.2 MBytes  5.45 Mbits/sec    6             sender
[  5]   0.00-101.93 sec  0.00 Bytes  0.00 bits/sec                  receiver
iperf3: interrupt - the client has terminated

On a separate console the log shows this while iperf3 is active:

Tue Nov  9 17:49:50 2021 kern.info kernel: [  903.051213] RX buffer overrun: status 101, mask: ffeff
Tue Nov  9 17:49:50 2021 kern.info kernel: [  903.057182] RX buffer overrun: status 101, mask: ffeff
Tue Nov  9 17:49:50 2021 kern.info kernel: [  903.063026] RX buffer overrun: status 1, mask: ffeff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  903.937196] RX buffer overrun: status 101, mask: ffcff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  903.943113] RX buffer overrun: status 1, mask: ffcff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  903.948754] RX buffer overrun: status 1, mask: ffcff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  903.954392] RX buffer overrun: status 101, mask: ffcff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  903.960362] RX buffer overrun: status 101, mask: ffcff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  903.966204] RX buffer overrun: status 1, mask: ffcff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  904.056387] RX buffer overrun: status 101, mask: ffeff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  904.062297] RX buffer overrun: status 1, mask: ffeff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  904.067934] RX buffer overrun: status 1, mask: ffeff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  904.073570] RX buffer overrun: status 101, mask: ffeff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  904.079537] RX buffer overrun: status 101, mask: ffeff
Tue Nov  9 17:49:51 2021 kern.info kernel: [  904.085384] RX buffer overrun: status 1, mask: ffeff

After terminating iperf3 the message changes and it remains as follows even though the high load situation is now gone:

Tue Nov  9 17:49:59 2021 kern.warn kernel: [  911.986238] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000c4, cur a70000c8
Tue Nov  9 17:49:59 2021 kern.warn kernel: [  911.996379] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000c4, cur a70000c8
Tue Nov  9 17:50:00 2021 kern.warn kernel: [  913.020755] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000cc, cur a70000d8
Tue Nov  9 17:50:00 2021 kern.warn kernel: [  913.030895] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000cc, cur a70000d8
Tue Nov  9 17:50:00 2021 kern.warn kernel: [  913.242054] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000d0, cur a70000d8
Tue Nov  9 17:50:00 2021 kern.warn kernel: [  913.258000] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000d4, cur a70000d8
Tue Nov  9 17:50:01 2021 kern.warn kernel: [  914.217759] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000e8, cur a70000ec
Tue Nov  9 17:50:01 2021 kern.warn kernel: [  914.227900] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000e8, cur a70000ec
Tue Nov  9 17:50:50 2021 kern.warn kernel: [  963.623829] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000c4, cur a70000c8
Tue Nov  9 17:50:50 2021 kern.warn kernel: [  963.633968] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000c4, cur a70000c8
Tue Nov  9 17:50:50 2021 kern.warn kernel: [  963.830442] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000cc, cur a70000d8
Tue Nov  9 17:50:50 2021 kern.warn kernel: [  963.840583] rtl838x-eth bb00a300.ethernet eth0: Ring contention: r: 0, last a70000cc, cur a70000d8

Traffic passing the switch from one port to another is not affected, it seems to pause traffic to eth0 for brief moments. To observe that a ping -A 192.168.xx.254 can be used.

It is surprising, that the device enters a state where it is loosing packets, even if high traffic stops and remains in this poor state.

Thank you for reading.

More information can be found at the following URL:
https://bugs.openwrt.org/index.php?do=details&task_id=4161

You are receiving this message because you have requested it from the Flyspray bugtracking system.  If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.