Messages

Issue: Lots of RX errors / missed packets when the NIC is connected to Arris ONT Calix 711GE ONT, connecting it to switch or computer gives no error. Swapping igc0/igc1 makes no difference. So many pkts are missed that it will frequently look like DNS issues, due to missed SYN-ACKs and long delays to connect to websites.

Code Select Expand

# sysctl dev.igc.1.mac_stats.missed_packets
dev.igc.1.mac_stats.missed_packets: 15467

# netstat -I igc1 | head 2
Name    Mtu Network         Address                                           Ipkts  Ierrs  Idrop    Opkts  Oerrs   Coll
igc1   1500 <Link#4>        64:62:66:XX:XX:XX                               1298933  15597      0   629757      0      0



Device info:

Code Select Expand

# dmesg | grep '\[1\] igc'
[1] igc0: <Intel(R) Ethernet Controller I226-V> mem 0x80400000-0x804fffff,0x80600000-0x80603fff at device 0.0 on pci2
[1] igc0: EEPROM V2.17-0 eTrack 0x80000303
[1] igc0: Using 1024 TX descriptors and 1024 RX descriptors
[1] igc0: Using 4 RX queues 4 TX queues
[1] igc0: Using MSI-X interrupts with 5 vectors
[1] igc0: Ethernet address: 64:62:66:XX:XX:XX
[1] igc0: netmap queues/slots: TX 4/1024, RX 4/1024
[1] igc1: <Intel(R) Ethernet Controller I226-V> mem 0x80700000-0x807fffff,0x80900000-0x80903fff at device 0.0 on pci4
[1] igc1: EEPROM V2.17-0 eTrack 0x80000303
[1] igc1: Using 1024 TX descriptors and 1024 RX descriptors
[1] igc1: Using 4 RX queues 4 TX queues
[1] igc1: Using MSI-X interrupts with 5 vectors
[1] igc1: Ethernet address: 64:62:66:XX:XX:XY
[1] igc1: netmap queues/slots: TX 4/1024, RX 4/1024

# pciconf -lV | grep igc
igc0@pci0:2:0:0:    class=0x020000 rev=0x04 hdr=0x00 vendor=0x8086 device=0x125c subvendor=0x8086 subdevice=0x0000
igc1@pci0:4:0:0:    class=0x020000 rev=0x04 hdr=0x00 vendor=0x8086 device=0x125c subvendor=0x8086 subdevice=0x0000

You will notice RSS enabled, I just enabled it for testing. It made no difference.


UPDATE: Inspired by another users' post, I picked up an old EdgeRoute, configured it as a dumb switch and put it in between opnsense and the OTN, RX errors dropped to zero
UPDATE2: updated one of the interfaces from V2.17-0 to V2.32-0 (2M). It did not fix the issue. Moved connection back to 2.17 with dumb switch
UPDATE3: after a couple days, family complained about transient connection issues. Although I can't see if there are RX errors on dumb switch interface there's none on opnsense interface, but lots of duplicate ACK just like before