sudo apt install ethtool
osmc@Vero:~$ sudo ethtool -a eth0
Pause parameters for eth0:
Autonegotiate: on
RX: off
TX: off
RX negotiated: off
TX negotiated: off
osmc@Vero:~$ sudo ethtool -S eth0 | grep -v 0$
NIC statistics:
mmc_rx_framecount_gb: 753
mmc_rx_octetcount_gb: 177162
mmc_rx_octetcount_g: 177162
mmc_rx_broadcastframe_g: 313
mmc_rx_multicastframe_g: 91
mmc_rx_64_octets_gb: 371
mmc_rx_65_to_127_octets_gb: 252
mmc_rx_128_to_255_octets_gb: 34
mmc_rx_512_to_1023_octets_gb: 8
mmc_rx_1024_to_max_octets_gb: 78
mmc_rx_unicast_g: 349
mmc_rx_ipc_intr_mask: 2147385342
mmc_rx_ipv4_nopay: 1
mmc_rx_ipv4_gd_octets: 149054
mmc_rx_ipv4_nopay_octets: 46
mmc_rx_ipv6_gd_octets: 368
mmc_rx_ipv6_gd: 4
mmc_rx_udp_gd: 273
mmc_rx_tcp_gd: 177
mmc_rx_icmp_gd: 4
mmc_rx_udp_gd_octets: 24254
mmc_rx_icmp_gd_octets: 208
threshold: 64
tx_pkt_n: 363
rx_pkt_n: 753
normal_irq_n: 521
rx_normal_irq_n: 516
napi_poll: 521
tx_normal_irq_n: 5
tx_clean: 712
tx_reset_ic_bit: 358
irq_rgmii_n: 1There are no errors. I suspect the switch to go into power-save mode and lower the interface speeds.
1 Like
Where is the switch?
OK, iptv.merge is not the problem.
ethtool:
~$ ~#
The following connections are open:
#0 client-session (t4 r0 i0/0 o0/0 fd 4/5 cc -1)
sudo ethtool -a eth0
Pause parameters for eth0:
Autonegotiate: on
RX: off
TX: off
RX negotiated: on
TX negotiated: on
osmc@osmc:~$ ~#
The following connections are open:
#0 client-session (t4 r0 i0/0 o0/0 fd 4/5 cc -1)
sudo ethtool -S eth0 | grep -v 0$
NIC statistics:
mmc_rx_framecount_gb: 5848871
mmc_rx_octetcount_gb: 3822715461
mmc_rx_octetcount_g: 3822715461
mmc_rx_broadcastframe_g: 58802
mmc_rx_multicastframe_g: 85172
mmc_rx_64_octets_gb: 342262
mmc_rx_65_to_127_octets_gb: 114397
mmc_rx_128_to_255_octets_gb: 18067
mmc_rx_256_to_511_octets_gb: 27211
mmc_rx_512_to_1023_octets_gb: 19405
mmc_rx_1024_to_max_octets_gb: 5327529
mmc_rx_unicast_g: 5704897
mmc_rx_ipc_intr_mask: 1073692671
mmc_rx_ipv4_gd: 5833486
mmc_rx_ipv4_nopay: 2364
mmc_rx_ipv4_gd_octets: 3714766891
mmc_rx_ipv4_nopay_octets: 108744
mmc_rx_udp_gd: 150779
mmc_rx_tcp_gd: 5682705
mmc_rx_icmp_gd: 2
mmc_rx_udp_gd_octets: 17456406
mmc_rx_tcp_gd_octets: 3580640693
mmc_rx_icmp_gd_octets: 72
threshold: 64
tx_pkt_n: 1098194
rx_pkt_n: 5848871
rx_normal_irq_n: 500497
tx_normal_irq_n: 25508
tx_clean: 570057
tx_reset_ic_bit: 1069812
irq_rgmii_n: 1
do I have one of the boxes with the dodgy ethernet? I seem to have exhausted everything else.
If it was from an early batch of Vero4K+ devices, then itās possible. When did you buy it?
It was december 2018
There is no switch between my router and the Vero.
I only have a switch between my router and my PC, from which I am running the SSH commands.
If your device is affected, then the TX direction will be affected only. It will always be affected and not after some time.
If a reboot fixes it, it canāt be a hardware issue.
Any idea why I only get the drops when running OSMC?
As said before, I get no drops on my Pi running plain Raspbian.
But when I installed OSMC on another SD card on my Pi, I got the same drops as on the Vero.
Nothing obvious that I can think of. The connection manager is different to Raspbian however. But I donāt think the drops are your issue.
Let it run for a while, and check if the drop counter is increasing. Usually the initial drops happens on a reboot when external devices still try to access it, but the endpoint does not exist anymore.
I have another question. If I run iperf3 without any additional parameters on the client side I get around 947 Mbit/s which is the maximum on Gbit LAN.
But if I run it on UDP protocol I only get aroung 70 Mbit/s. Is this normal?
Command on my Windows machine:
.\iperf3.exe -s
TCP test:
osmc@Vero:~$ iperf3 -c 192.168.178.21
Connecting to host 192.168.178.21, port 5201
[ 4] local 192.168.178.25 port 33875 connected to 192.168.178.21 port 5201
[ ID] Interval Transfer Bandwidth Retr Cwnd
[ 4] 0.00-1.00 sec 113 MBytes 943 Mbits/sec 0 269 KBytes
[ 4] 1.00-2.00 sec 113 MBytes 947 Mbits/sec 0 269 KBytes
[ 4] 2.00-3.00 sec 113 MBytes 946 Mbits/sec 0 269 KBytes
[ 4] 3.00-4.00 sec 113 MBytes 944 Mbits/sec 0 269 KBytes
[ 4] 4.00-5.00 sec 113 MBytes 947 Mbits/sec 0 269 KBytes
[ 4] 5.00-6.00 sec 113 MBytes 946 Mbits/sec 0 269 KBytes
[ 4] 6.00-7.00 sec 113 MBytes 946 Mbits/sec 0 269 KBytes
[ 4] 7.00-8.00 sec 113 MBytes 947 Mbits/sec 0 269 KBytes
[ 4] 8.00-9.00 sec 113 MBytes 947 Mbits/sec 0 269 KBytes
[ 4] 9.00-10.00 sec 113 MBytes 947 Mbits/sec 0 269 KBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth Retr
[ 4] 0.00-10.00 sec 1.10 GBytes 946 Mbits/sec 0 sender
[ 4] 0.00-10.00 sec 1.10 GBytes 945 Mbits/sec receiver
iperf Done.
UDP test:
osmc@Vero:~$ iperf3 -c 192.168.178.21 -u -b 1000M
Connecting to host 192.168.178.21, port 5201
[ 4] local 192.168.178.25 port 40386 connected to 192.168.178.21 port 5201
[ ID] Interval Transfer Bandwidth Total Datagrams
[ 4] 0.00-1.00 sec 7.61 MBytes 63.8 Mbits/sec 974
[ 4] 1.00-2.00 sec 8.57 MBytes 71.9 Mbits/sec 1097
[ 4] 2.00-3.00 sec 8.43 MBytes 70.7 Mbits/sec 1079
[ 4] 3.00-4.00 sec 8.59 MBytes 72.0 Mbits/sec 1099
[ 4] 4.00-5.00 sec 8.31 MBytes 69.7 Mbits/sec 1064
[ 4] 5.00-6.00 sec 8.59 MBytes 72.1 Mbits/sec 1100
[ 4] 6.00-7.00 sec 8.41 MBytes 70.5 Mbits/sec 1076
[ 4] 7.00-8.00 sec 8.58 MBytes 72.0 Mbits/sec 1098
[ 4] 8.00-9.00 sec 8.51 MBytes 71.4 Mbits/sec 1089
[ 4] 9.00-10.00 sec 8.66 MBytes 72.7 Mbits/sec 1109
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth Jitter Lost/Total Datagrams
[ 4] 0.00-10.00 sec 84.3 MBytes 70.7 Mbits/sec 0.053 ms 0/10783 (0%)
[ 4] Sent 10783 datagrams
iperf Done.Any idea whatās going on?
Did you test with iperf?
me? Iām not following - this whole thread is about my iperf results dropping from 1Gig levels to around 150-200mb in (now) a short period of time. on both tx and rx if thatās what you mean.
The only thing that will fix the speeds is a reboot, or now we find, disabling the deviceās wired ethernet and reenabling. Pulling cables does nothing, rebooting all the other network switches/routers etc does nothing. Itās pretty clear the device itself is slowing down for some reason?
I wouldnāt get too hung up on those UDP figures, which I donāt believe are relevant to your problem. I have a non-plus Vero4K and see similar UDP figures, though the reverse figures seem fine. It looks to me like an iperf3-specific issue.
Iām interested to see if the speed degradation you see occurs rapidly or gradually over time.
If cron isnāt already installed, please install it and then run crontab -e. (It might ask you which editor you want to use.) Then add/save this line:
*/15 * * * * /usr/bin/iperf3 -V -c 192.168.178.21 -i 0 --logfile /home/osmc/ip3.out >/tmp/ip3.log 2>&1
This will run an iperf3 every 15 minutes. Let it run overnight, preferably starting after a reboot. Assuming that thereās been a slowdown, you can upload the file with paste-log /home/osmc/ip3.out.