interface errors

mervincm

Any Ideas on troubleshooting this? Strangely I have multiple inbound, but zero outbound.
They increment continually.

Both of these are on an dual port intel 10gig card. ix0/ix1

stephenw10

Check the mac stats in the sysctl output. That will show you what type of errors they are.
See: https://redmine.pfsense.org/issues/12904

Steve

mervincm

@stephenw10 thx so much for the quick response

sysctl -a | grep checksum_errs
dev.ix.1.mac_stats.checksum_errs: 1673
dev.ix.0.mac_stats.checksum_errs: 102

mervincm

@mervincm
22.05
hw checksum Offloading, hardware TCP segmentation offloading, hw large recieve offloading, are disabled

I do not have a switch on the WAN interface, but I do use a SFP+ transceiver to connect to my ISP device @ 10gig

I use a twinax cable from my LAN interface to my Switch @ 10gig

edit I mistakenly said my LAN port post transciever was @2.5g when it is actually at 10G

stephenw10

Mmm, so mostly not checksum errors in your case. There are a number of error types listed in the full mac_stats output there. Check to see where most are.

Steve

mervincm

@stephenw10

I re-booted in-between the two posts to disable hardware checksum offload.

Mine are infact all checksum errors.
I included the first screenshot to show the numbers are not insignificant. Or perhaps they are and they are "normal" and I can ignore these. I know that before I upgraded the NIC to support this new ISP service I had very few errors, often none when I checked.

sysctl -a | grep checksum_errs
dev.ix.1.mac_stats.checksum_errs: 3932
dev.ix.0.mac_stats.checksum_errs: 201

stephenw10

Ah, OK. That might just be what there is on your line. Though it's odd having that on the LAN.

Check the output of: ifconfig -vvvm ix0 and see what other offloading options are enabled.

Steve

mervincm

@stephenw10

ifconfig -vvvm ix0
ix0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
        description: WAN
        options=8138b8<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,WOL_UCAST,WOL_MCAST,WOL_MAGIC,VLAN_HWFILTER>
        capabilities=f53fbb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,WOL_UCAST,WOL_MCAST,WOL_MAGIC,VLAN_HWFILTER,VLAN_HWTSO,NETMAP,RXCSUM_IPV6,TXCSUM_IPV6>
        ether 38:ea:a7:33:ad:10
        inet6 fe80::3aea:a7ff:fe33:ad10%ix0 prefixlen 64 scopeid 0x1
        inet 206.75.30.245 netmask 0xfffffc00 broadcast 206.75.31.255
        media: Ethernet autoselect (10Gbase-Twinax <full-duplex,rxpause,txpause>)
        status: active
        supported media:
                media autoselect
                media 10Gbase-Twinax
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        plugged: SFP/SFP+/SFP28 Unknown (RJ45)
        vendor: MikroTik PN: S+RJ10 SN: 815D0301DD22 DATE: 2018-06-28

-vvvm ix1
ix1: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
        description: LAN
        options=8138b8<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,WOL_UCAST,WOL_MCAST,WOL_MAGIC,VLAN_HWFILTER>
        capabilities=f53fbb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,WOL_UCAST,WOL_MCAST,WOL_MAGIC,VLAN_HWFILTER,VLAN_HWTSO,NETMAP,RXCSUM_IPV6,TXCSUM_IPV6>
        ether 38:ea:a7:33:ad:11
        inet6 fe80::3aea:a7ff:fe33:ad11%ix1 prefixlen 64 scopeid 0x2
        inet 10.0.0.1 netmask 0xfffff800 broadcast 10.0.7.255
        inet 10.0.0.5 netmask 0xfffff800 broadcast 10.0.7.255
        media: Ethernet autoselect (10Gbase-Twinax <full-duplex,rxpause,txpause>)
        status: active
        supported media:
                media autoselect
                media 10Gbase-Twinax
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        plugged: SFP/SFP+/SFP28 Unknown (Copper pigtail)
        vendor: ADDON PN: AOCABSFP+SFP+3M SN: ADD3M02229 DATE: 2008-06-25

stephenw10

Hmm well you have VLAN_HWCSUM enabled there. Do you have VLANs on those NICs?

mervincm

@stephenw10
Nope

netstat -i
Name    Mtu Network       Address              Ipkts Ierrs Idrop    Opkts Oerrs  Coll
ix0    1500 <Link#1>      38:ea:a7:33:ad:10 12760934  8455     0  5176975     0     0
ix0       - fe80::%ix0/64 fe80::3aea:a7ff:f        0     -     -        2     -     -
ix0       - 206.75.28.0/2 d206-75-30-245.ab   576512     -     -   151095     -     -
ix1    1500 <Link#2>      38:ea:a7:33:ad:11 12095522 182243     0 22065569     0     0
ix1       - fe80::%ix1/64 fe80::3aea:a7ff:f        0     -     -        0     -     -
ix1       - 10.0.0.0/21   rtr                6827253     -     - 10006757     -     -
ix1       - 10.0.0.0/21   ha_proxy_vip           297     -     -        0     -     -
em0*   1500 <Link#3>      fc:4d:d4:f5:32:43        0     0     0        0     0     0
enc0*  1536 <Link#4>      enc0                     0     0     0        0     0     0
lo0   16384 <Link#5>      lo0                 439806     0     0   439806     0     0
lo0       - localhost     localhost                0     -     -        0     -     -
lo0       - fe80::%lo0/64 fe80::1%lo0              0     -     -        0     -     -
lo0       - your-net      localhost           439806     -     -   439806     -     -
pfsyn  1500 <Link#6>      pfsync0                  0     0     0        0     0     0
pflog 33160 <Link#7>      pflog0                   0     0     0   172529     0     0
tun_w  1420 <Link#8>      tun_wg0             282754    12     0   387108    24     0
tun_w     - 10.1.15.0/24  10.1.15.1             2959     -     -     2782     -     -

stephenw10

Hmm, well, I'm not really sure what causes those errors. I agree I wouldn't expect to see any really, especially on an internal interface.
I assume you're not actually seeing any problems with the traffic using those NICs?

Steve

mervincm

@stephenw10 I don’t have a problem I am chasing down. I didn’t see these issues when I ran ESXI so I think the NIC is functional. I have fab cooling on it so it’s not overheating. Perhaps I should order a replacement NIC, I only need 2.5G as my ISP is only 1.5. I will see if there are any dual port Intel 2.5G NIC for a fair price. Thanks for the assistance.

mervincm

@mervincm

Moving to 23.01-RELEASE has completely cleaned this up.

This can be closed.

stephenw10

Nice. Thanks for following up.