So to close the loop on mail list, after offline discussions, this has to do
with configuring bonding and VLAN's. This was not an issue with the drivers or
the HW.
Thanks.
Cheers,
John
From: Jekels, Donny [mailto:[email protected]]
Sent: Wednesday, January 30, 2013 9:47 AM
To: Ronciak, John
Cc: [email protected]
Subject: Re: [E1000-devel] ARPING does not work
John,
Here is what I am talking about, all basic network functionality works.
we also have set the arp_announce in the kernel settings to 1 that is supposed
to fix this behavior and it does but not on these NIC's
the IP that I used for this testing is 10.8.198.250
when I plumb it up and arping from it does not respond to pings.
it ONLY starts working after I initate ping -I 10.8.198.250 10.11.11.11
then the IP on the interface works.
BUT arping fails completely
djekels@postgres19:~$ ip route
10.33.0.0/24<http://10.33.0.0/24>
nexthop via 10.4.42.1 dev eth4 weight 1
nexthop via 10.4.42.1 dev eth5 weight 1
nexthop via 10.4.42.1 dev eth6 weight 1
nexthop via 10.4.42.1 dev eth7 weight 1
10.8.198.0/24<http://10.8.198.0/24> dev vlan198 proto kernel scope link src
10.8.198.119
10.8.199.0/24<http://10.8.199.0/24> dev vlan199 proto kernel scope link src
10.8.199.39
10.32.0.0/24<http://10.32.0.0/24>
nexthop via 10.4.42.1 dev eth4 weight 1
nexthop via 10.4.42.1 dev eth5 weight 1
nexthop via 10.4.42.1 dev eth6 weight 1
nexthop via 10.4.42.1 dev eth7 weight 1
10.4.42.0/24<http://10.4.42.0/24> dev eth4 proto kernel scope link src
10.4.42.96
10.4.42.0/24<http://10.4.42.0/24> dev eth5 proto kernel scope link src
10.4.42.97
10.4.42.0/24<http://10.4.42.0/24> dev eth6 proto kernel scope link src
10.4.42.98
10.4.42.0/24<http://10.4.42.0/24> dev eth7 proto kernel scope link src
10.4.42.99
172.23.16.0/23<http://172.23.16.0/23> dev vlan4000 proto kernel scope link
src 172.23.16.211
172.23.0.0/16<http://172.23.0.0/16> via 172.23.16.1 dev vlan4000
default via 10.8.198.1 dev vlan198
djekels@postgres19:~$
djekels@postgres19:~$ ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8<http://127.0.0.1/8> scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc mq master bond0
state UP qlen 1000
link/ether bc:30:5b:f0:a3:80 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc mq master bond0
state UP qlen 1000
link/ether bc:30:5b:f0:a3:80 brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
link/ether bc:30:5b:f0:a3:82 brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
link/ether bc:30:5b:f0:a3:83 brd ff:ff:ff:ff:ff:ff
6: eth4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc mq state UP qlen 1000
link/ether a0:36:9f:08:09:c8 brd ff:ff:ff:ff:ff:ff
inet 10.4.42.96/24<http://10.4.42.96/24> brd 10.4.42.255 scope global eth4
inet6 fe80::a236:9fff:fe08:9c8/64 scope link
valid_lft forever preferred_lft forever
7: eth5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc mq state UP qlen 1000
link/ether a0:36:9f:08:09:c9 brd ff:ff:ff:ff:ff:ff
inet 10.4.42.97/24<http://10.4.42.97/24> brd 10.4.42.255 scope global eth5
inet6 fe80::a236:9fff:fe08:9c9/64 scope link
valid_lft forever preferred_lft forever
8: eth6: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc mq state UP qlen 1000
link/ether a0:36:9f:08:09:ca brd ff:ff:ff:ff:ff:ff
inet 10.4.42.98/24<http://10.4.42.98/24> brd 10.4.42.255 scope global eth6
inet6 fe80::a236:9fff:fe08:9ca/64 scope link
valid_lft forever preferred_lft forever
9: eth7: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc mq state UP qlen 1000
link/ether a0:36:9f:08:09:cb brd ff:ff:ff:ff:ff:ff
inet 10.4.42.99/24<http://10.4.42.99/24> brd 10.4.42.255 scope global eth7
inet6 fe80::a236:9fff:fe08:9cb/64 scope link
valid_lft forever preferred_lft forever
10: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue
state UP
link/ether bc:30:5b:f0:a3:80 brd ff:ff:ff:ff:ff:ff
inet6 fe80::be30:5bff:fef0:a380/64 scope link
valid_lft forever preferred_lft forever
11: vlan4000@bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc
noqueue state UP
link/ether bc:30:5b:f0:a3:80 brd ff:ff:ff:ff:ff:ff
inet 172.23.16.211/23<http://172.23.16.211/23> brd 172.23.17.255 scope
global vlan4000
inet6 fe80::be30:5bff:fef0:a380/64 scope link
valid_lft forever preferred_lft forever
12: vlan198@bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc
noqueue state UP
link/ether bc:30:5b:f0:a3:80 brd ff:ff:ff:ff:ff:ff
inet 10.8.198.119/24<http://10.8.198.119/24> brd 10.8.198.255 scope global
vlan198
inet 10.8.198.134/24<http://10.8.198.134/24> brd 10.8.198.255 scope global
secondary vlan198
inet 10.8.198.167/24<http://10.8.198.167/24> brd 10.8.198.255 scope global
secondary vlan198
inet 10.8.198.250/24<http://10.8.198.250/24> brd 10.8.198.255 scope global
secondary vlan198
inet6 fe80::be30:5bff:fef0:a380/64 scope link
valid_lft forever preferred_lft forever
13: vlan199@bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc
noqueue state UP
link/ether bc:30:5b:f0:a3:80 brd ff:ff:ff:ff:ff:ff
inet 10.8.199.39/24<http://10.8.199.39/24> brd 10.8.199.255 scope global
vlan199
inet 10.8.199.105/24<http://10.8.199.105/24> brd 10.8.199.255 scope global
secondary vlan199
inet 10.8.199.108/24<http://10.8.199.108/24> brd 10.8.199.255 scope global
secondary vlan199
inet6 fe80::be30:5bff:fef0:a380/64 scope link
valid_lft forever preferred_lft forever
djekels@postgres19:~$ netstat -s
Ip:
71751543 total packets received
166 with invalid addresses
0 forwarded
0 incoming packets discarded
71706086 incoming packets delivered
75075912 requests sent out
10 dropped because of missing route
Icmp:
3645 ICMP messages received
6 input ICMP message failed.
ICMP input histogram:
destination unreachable: 1716
echo requests: 1851
echo replies: 78
56066 ICMP messages sent
0 ICMP messages failed
ICMP output histogram:
destination unreachable: 53169
echo request: 1046
echo replies: 1851
IcmpMsg:
InType0: 78
InType3: 1716
InType8: 1851
OutType0: 1851
OutType3: 53169
OutType8: 1046
Tcp:
708305 active connections openings
3975 passive connection openings
431 failed connection attempts
33564 connection resets received
14 connections established
68956581 segments received
72313084 segments send out
24468 segments retransmited
0 bad segments received.
30446 resets sent
Udp:
2671621 packets received
74239 packets to unknown port received.
0 packet receive errors
2682294 packets sent
UdpLite:
TcpExt:
396 resets received for embryonic SYN_RECV sockets
1075 packets pruned from receive queue because of socket buffer overrun
174110 TCP sockets finished time wait in fast timer
93 time wait sockets recycled by time stamp
1547391 delayed acks sent
47 delayed acks further delayed because of locked socket
Quick ack mode was activated 887 times
723398 packets directly queued to recvmsg prequeue.
5496 bytes directly in process context from backlog
4362730 bytes directly received in process context from prequeue
49343700 packet headers predicted
17131 packets header predicted and directly queued to user
1638076 acknowledgments not containing data payload received
31754543 predicted acknowledgments
181 times recovered from packet loss by selective acknowledgements
52 congestion windows recovered without slow start by DSACK
1023 congestion windows recovered without slow start after partial ack
191 TCP data loss events
TCPLostRetransmit: 15
114 timeouts after SACK recovery
2 timeouts in loss state
314 fast retransmits
15 forward retransmits
54 retransmits in slow start
12748 other TCP timeouts
13 SACK retransmits failed
385169 packets collapsed in receive queue due to low socket buffer
874 DSACKs sent for old packets
236 DSACKs received
8 connections reset due to unexpected data
29985 connections reset due to early user close
91 connections aborted due to timeout
TCPDSACKIgnoredOld: 168
TCPDSACKIgnoredNoUndo: 9
TCPSackShifted: 259
TCPSackMerged: 503
TCPSackShiftFallback: 530
IpExt:
InBcastPkts: 37803
InOctets: -1264740945
OutOctets: -592937989
InBcastOctets: 9434134
On Wed, Jan 30, 2013 at 11:34 AM, Ronciak, John
<[email protected]<mailto:[email protected]>> wrote:
So the NIC's are sending and receiving packets. What is the stack doing
('netstat -s')? My guess is that it's not doing the right thing with broadcast
packets. Do you have a resolv.conf file configured to do what you want? Maybe
it's not correct?
Cheers,
John
From: Jekels, Donny [mailto:[email protected]<mailto:[email protected]>]
Sent: Wednesday, January 30, 2013 9:10 AM
To: Ronciak, John
Cc: [email protected]<mailto:[email protected]>
Subject: Re: [E1000-devel] ARPING does not work
John,
The NIC works for TCP and UDP sessions however no broadcast and arp seems to
work.
we move IP's around between servers and part of this move is to garp the source
IP so the switches can update their mac tables.
without this functionality the NIC's are useless to us.
please help
root@postgres19:~# ethtool -S eth7
NIC statistics:
rx_packets: 11
tx_packets: 6
rx_bytes: 704
tx_bytes: 492
rx_broadcast: 11
tx_broadcast: 0
rx_multicast: 0
tx_multicast: 6
multicast: 0
collisions: 0
rx_crc_errors: 0
rx_no_buffer_count: 0
rx_missed_errors: 0
tx_aborted_errors: 0
tx_carrier_errors: 0
tx_window_errors: 0
tx_abort_late_coll: 0
tx_deferred_ok: 0
tx_single_coll_ok: 0
tx_multi_coll_ok: 0
tx_timeout_count: 0
rx_long_length_errors: 0
rx_short_length_errors: 0
rx_align_errors: 0
tx_tcp_seg_good: 0
tx_tcp_seg_failed: 0
rx_flow_control_xon: 0
rx_flow_control_xoff: 0
tx_flow_control_xon: 0
tx_flow_control_xoff: 0
rx_long_byte_count: 704
tx_dma_out_of_sync: 0
lro_aggregated: 0
lro_flushed: 0
lro_recycled: 0
tx_smbus: 0
rx_smbus: 0
dropped_smbus: 0
os2bmc_rx_by_bmc: 0
os2bmc_tx_by_bmc: 0
os2bmc_tx_by_host: 0
os2bmc_rx_by_host: 0
rx_errors: 0
tx_errors: 0
tx_dropped: 0
rx_length_errors: 0
rx_over_errors: 0
rx_frame_errors: 0
rx_fifo_errors: 0
tx_fifo_errors: 0
tx_heartbeat_errors: 0
tx_queue_0_packets: 6
tx_queue_0_bytes: 468
tx_queue_0_restart: 0
rx_queue_0_packets: 11
rx_queue_0_bytes: 660
rx_queue_0_drops: 0
rx_queue_0_csum_err: 0
rx_queue_0_alloc_failed: 0
root@postgres19:~# ethtool -i eth7
driver: igb
version: 4.1.2
firmware-version: 1.6, 0x80000816
bus-info: 0000:04:00.3
root@postgres19:~# arping 10.11.12.13
ARPING 10.11.12.13
On Wed, Jan 30, 2013 at 11:05 AM, Ronciak, John
<[email protected]<mailto:[email protected]>> wrote:
Hi Donny,
Sorry to hear you are having problems. This most likely is not a NIC/driver
issue. Something is mostly not configured correctly on these systems. Did you
use ethtool to look at the HW stats of the interfaces you are trying to ping
out of? If not, please do so. Also, are each port/interface on a different
subnet? The stack does not do well (without extra configuration) with multiple
ports on the same subnet.
The reason Debian Squeeze doesn't have support is that they didn't pick up an
updated driver before they released it. This is not a problem with the Intel
drivers, it's a timing thing for Debian releases and how they update kernels
and drivers.
Cheers,
John
> -----Original Message-----
> From: Jekels, Donny [mailto:[email protected]<mailto:[email protected]>]
> Sent: Tuesday, January 29, 2013 8:35 PM
> To:
> [email protected]<mailto:[email protected]>
> Subject: [E1000-devel] ARPING does not work
>
> We have a handful of these Intel quad port NIC with the drive 4.1.7 and
> arping does not work on these NIC's.
>
> we run debian stock squeeze and had to compile igb driver from intel to
> get the NIC ordering correct.
>
> now arping is also not working.
>
> can you give me some advise how do I continue having this issue
> resolved?
>
> thanks,
> Donny
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_jan
_______________________________________________
E1000-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit
http://communities.intel.com/community/wired
