Solved Intel (igb) Quad-Port (82576) crashing at load

_martin said:
Limit your RAM to one module during tests, it will speed up the checks too.
Under-power could be a problem if the PSU is weak. You could put at least wattmeter in line to see if the power drawn is maybe too much for that PSU.

Some other things I'd do:
Test NIC somewhere else
Test different NIC in that PCIe slot

If you are under warranty you could try to RMA that NIC too.

An obvious question but still - have you checked the /var/log/messages for doublefault messages, etc.? Both on FreeBSD and Linux.
Click to expand...
I have only one module. Up to now (still running) no error.

I would assume power is not an issue. I ran the same test with the onBoard NIC (also connected via PCIe lanes) without issues. At least for me it's very difficult to test. I read other have issues with those Intel NICs with the AMD CPUs on embedded systems like this Fujitsu S930. So I assume the Intel card doesnt like the AMD CPU.

I got the Intel card for free now. I wanted to send it back but they just refunded the money. Maybe Im not the first one to send it back ;) I have another system with PCIe X1 slot, but this wont work of course, so I cant really test it somewhere else.

/var/log/messages was not helpful. The only thing I found in dmesg was some error which stated "ignored". Further research pointed to a discussion that somebody should "remove this message" and rewrite the driver.
 
Yes, I will replace the network card with maybe a Broadcom. This seems to work and is similar priced.

Thanks for your help and suggestions everybody!
 
Many second hand hardware is sold by disassemblers of systems. They buy up entire (sometimes broken) systems, disassemble them and sell the parts.
They generally don't test them and their policy is simple: if the customer says it's broken, refund the customer. It's cheaper to let the customers do the testing for them.
 
Ben said:
For later reference: No solution found, assuming a damaged NIC OR an incompatibility between AMD processor and this Intel chipset.
Click to expand...
I normally don't revive old threads but I have the Dell model of the 82576 quad port nic with a Fujitsu S920 and have the same issues. Adding to this thread for future reference. No crash data whatsoever. I tried to load it in Windows 10/11 and Ubuntu 22.04 with no issues in my tower so the card isn't damaged. Probably just an incompatibility with the AMD.
 
I have 2pcs Intel 82575EB Gigabit Network cards with 2 ports + AMD Ryzen 7 5700G + ASUS TUF GAMING B550-PLUS. FreeBSD 14.3 releng/14.3
It's constantly crashing in iflib at /usr/src/sys/net/iflib.c:2917

sh: 
57              __asm("movq %%gs:%P1,%0" : "=r" (td) : "n" (offsetof(struct pcpu,
(kgdb) #0  __curthread () at /usr/src/sys/amd64/include/pcpu_aux.h:57
        td = <optimized out>
#1  doadump (textdump=<optimized out>) at /usr/src/sys/kern/kern_shutdown.c:405
        error = 0
        coredump = <optimized out>
#2  0xffffffff808292cb in kern_reboot (howto=260)
    at /usr/src/sys/kern/kern_shutdown.c:523
        once = 0
#3  0xffffffff808297c9 in vpanic (fmt=0xffffffff80d912d4 "%s",
    ap=ap@entry=0xfffffe0068bc6bd0) at /usr/src/sys/kern/kern_shutdown.c:967
        buf = "page fault", '\000' <repeats 245 times>
        __pc = 0x0
        __pc = 0x0
        __pc = 0x0
        other_cpus = {__bits = {65534, 0 <repeats 15 times>}}
        td = 0xfffff80001c47740
        bootopt = <unavailable>
        newpanic = <optimized out>
#4  0xffffffff80829603 in panic (fmt=<unavailable>)
    at /usr/src/sys/kern/kern_shutdown.c:891
        ap = {{gp_offset = 16, fp_offset = 48,
            overflow_arg_area = 0xfffffe0068bc6c00,
            reg_save_area = 0xfffffe0068bc6ba0}}
#5  0xffffffff80c87a1a in trap_fatal (frame=<optimized out>,
    eva=<optimized out>) at /usr/src/sys/amd64/amd64/trap.c:1000
        type = <optimized out>
        handled = <optimized out>
#6  0xffffffff80c87a1a in trap_pfault (frame=0xfffffe0068bc6c50,
    usermode=false, signo=<optimized out>, ucode=<optimized out>)
        __pc = 0x0
        __pc = 0x0
        __pc = 0x0
        td = <optimized out>
        p = <optimized out>
        eva = <optimized out>
        map = <optimized out>
        ftype = <optimized out>
        rv = <optimized out>
#7  <signal handler called>
No locals.
#8  iflib_check_lro_possible (m=0xfffff80228fad200,
    v4_forwarding=<optimized out>, v6_forwarding=<optimized out>)
    at /usr/src/sys/net/iflib.c:2917
        eh = 0xfffff0005ca8c800
#9  iflib_rxeof (rxq=rxq@entry=0xfffff80002837000, budget=<optimized out>)
    at /usr/src/sys/net/iflib.c:3057
        ri = <optimized out>
        ctx = 0xfffff800027dec00
        sctx = <optimized out>
        scctx = 0xfffff800027deca8
        retval = <optimized out>
        v6_forwarding = <optimized out>
        v4_forwarding = <optimized out>
        lro_possible = false
        ifp = 0xfffff80002848000
        mt = 0xfffff8028318cb00
        mh = 0x0
        rx_bytes = <optimized out>
        rx_pkts = <optimized out>
        cidxp = 0xfffff800027ba600
        avail = 0
        saved_vnet = 0x0
        budget_left = <optimized out>
        err = <optimized out>
        m = 0xfffff80228fad200
        i = <optimized out>
        fl = <optimized out>
        mf = 0xfffff8028318cb00
        lro_enabled = <optimized out>

Who is interested in fixing this issue I can provide several crash dumps.
 
You must log in or register to reply here.
Back
Top