Bug 150557 - [igb] igb0: Watchdog timeout -- resetting
Summary: [igb] igb0: Watchdog timeout -- resetting
Status: Closed Feedback Timeout
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 8.1-STABLE
Hardware: Any Any
: Normal Affects Only Me
Assignee: freebsd-net mailing list
URL:
Keywords: IntelNetworking
Depends on:
Blocks:
 
Reported: 2010-09-14 16:00 UTC by Slad
Modified: 2015-09-15 17:47 UTC (History)
2 users (show)

See Also:


Attachments
file.txt (25.58 KB, text/plain)
2010-09-14 16:00 UTC, Slad
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Slad 2010-09-14 16:00:03 UTC
There are the following server: MB X58A-UD3R, CPU Intel (R) Core (TM) i7 950, and 2 network cards E1G42ET. 2 ports on each board are grouped together in the lagg: lagg0 and lagg1

lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
        options=1bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4>
        ether 00:1b:21:5f:9d:6c
        inet 1.1.1.1 netmask 0xffffffc0 broadcast 1.1.1.63
        media: Ethernet autoselect
        status: active
        laggproto lacp
        laggport: igb1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
        laggport: igb0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>

lagg1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
        options=1bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4>
        ether 00:1b:21:6e:9e:cc
        inet 2.2.2.26 netmask 0xfffffffc broadcast 2.2.2.27
        media: Ethernet autoselect
        status: active
        laggproto lacp
        laggport: igb3 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
        laggport: igb2 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>

lagg0 connected to the same server is running in bridge mode, lagg1 connected to the Cisco 6509.

At high load - 1,7 Gbit / s start error:
igb0: Watchdog timeout -- resetting
igb0: Queue(2) tdh = 214, hw tdt = 438
igb0: TX(2) desc avail = 800,Next TX to Clean = 214
igb0: link state changed to DOWN
igb0: link state changed to UP

#cat /boot/loader.conf
hw.em.rxd=4096
hw.em.txd=4096
net.inet.tcp.tcbhashsize=4096
#hw.igb.rxd=2048
#hw.igb.txd=2048
hw.igb.lro.enable=0
hw.igb.num_queues=0

#netstat -w1 -h
            input        (Total)           output
   packets  errs idrops      bytes    packets  errs      bytes colls
      1.0M     0     0       694M       1.0M     0       737M     0
      1.1M  5.2K     0       716M       1.1M     0       721M     0
      1.1M     0     0       722M       1.1M     0       730M     0
      1.0M     0     0       698M       1.0M     0       730M     0
      1.1M  1.7K     0       726M       1.1M     0       729M     0
      1.0M     0     0       694M       1.0M     0       786M     0
      1.1M  1.5K     0       719M       1.1M     0       720M     0

top -SHI
last pid: 51590;  load averages:  1.53,  1.56,  1.44                                                 up 4+10:12:10  22:39:29
196 processes: 29 running, 115 sleeping, 52 waiting
CPU:  0.7% user,  0.0% nice, 42.1% system, 35.9% interrupt, 21.4% idle
Mem: 336M Active, 267M Inact, 724M Wired, 200K Cache, 418M Buf, 2621M Free
Swap: 4096M Total, 4096M Free

Fix: I do not know.

Patch attached with submission follows:
How-To-Repeat: Generate a large flow of traffic.
Comment 1 Mark Linimon freebsd_committer freebsd_triage 2010-09-20 08:09:01 UTC
Responsible Changed
From-To: freebsd-amd64->freebsd-kern

This does not sound amd64-specific.
Comment 2 Mark Linimon freebsd_committer freebsd_triage 2010-09-20 16:02:45 UTC
Responsible Changed
From-To: freebsd-kern->freebsd-bugs

Fix late-night brain-o.
Comment 3 Mark Linimon freebsd_committer freebsd_triage 2010-09-26 21:55:21 UTC
Responsible Changed
From-To: freebsd-bugs->freebsd-net

Over to maintainer(s).
Comment 4 Sean Bruno freebsd_committer 2015-08-04 15:23:47 UTC
Does this still happen for you with 10.2r?
Comment 5 Sean Bruno freebsd_committer 2015-09-15 17:47:01 UTC
Timing out this issue on 8.1 stable.  Please reopen this issue or open a new ticket if this is is happening on supported releases.