Hello, Freebsd-net. It's me again, as problem is not solved and no "clear" answer was received.
em0 NIC on my storage server hangs every several (2-3) days.
Symptoms are simple: no packets can be send, mbufs are overfilled, "No
buf space to send" error for any program.
Configuration now is VERY BASIC: no polling, no sysctls or
loader.conf tunables AT ALL. No jumbo frames.
nic doesn't show any "Watchdog timeout" / "resetting" messages.
Driver from "em driver, 82574L chip, and possibly ASPM" thread
doesn't help, really: it seems, that it decrease frequincy of hangs,
but doesn't eliminate them, but I can not say for sure, may be
frequency change is only illusion, as it random process.
I've added diagnostic patch from Michael Tuxen.
System is cvsupped and built Mon Feb 21, it is FreeBSD 8-STABLE
(RELENG_8).
Hardware is:
em0@pci0:0:25:0: class=0x020000 card=0x82681043 chip=0x10bd8086 rev=0x02
hdr=0x00
vendor = 'Intel Corporation'
device = 'Intel 82566DM Gigabit Ethernet Adapter (82566DM)'
class = network
subclass = ethernet
bar [10] = type Memory, range 32, base 0xfeb40000, size 131072, enabled
bar [14] = type Memory, range 32, base 0xfeb7a000, size 4096, enabled
bar [18] = type I/O Port, range 32, base 0xe880, size 32, enabled
cap 01[c8] = powerspec 2 supports D0 D3 current D0
cap 05[d0] = MSI supports 1 message, 64 bit
cap 09[e0] = vendor (length 6) Intel cap 2 version 0
Output of ifconfig, vmstat -m, netstat -m, top -Snd 1, sysctl
dev.em.0 is attached.
Interesting part of diagnostic sysctls:
dev.em.0.queue0.rxd_head: 896
dev.em.0.queue0.rxd_tail: 895
dev.em.0.queue0.rx_irq: 0
dev.em.0.queue0.rx_nxt_refresh: 896
dev.em.0.queue0.rx_nxt_check: 896
--
// Black Lion AKA Lev Serebryakov <[email protected]>
em0.log
Description: Binary data
_______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[email protected]"
