On Fri, Oct 7, 2011 at 8:13 AM, Ronciak, John <[email protected]> wrote:
> The reordering this occurring due to the retransmissions that are happening.
>> 2165 segments retransmitted
>
> Since these packets were dropped the other packets look like they are out of
> order which really isn't the case.
That makes sense, although when I captured traffic on both the
transmitter and receiver I often only saw a single copy of the packet
on the transmitter, which is why it looks like reordering to me. I
don't know whether what I captured was a representative sample though.
> What is causing the drops in another question and should be looked into. If
> you get us the device HW stats using 'ethtool -s p1p1' we might be able to
> see what's happening on your system.
On the transmitter:
# ethtool -S p1p1
NIC statistics:
rx_packets: 911117
tx_packets: 1758986
rx_bytes: 60151038
tx_bytes: 70692045072
rx_pkts_nic: 911117
tx_pkts_nic: 48750255
rx_bytes_nic: 63795506
tx_bytes_nic: 73988469882
lsc_int: 4
tx_busy: 0
non_eop_descs: 0
rx_errors: 0
tx_errors: 0
rx_dropped: 0
tx_dropped: 0
multicast: 0
broadcast: 0
rx_no_buffer_count: 0
collisions: 0
rx_over_errors: 0
rx_crc_errors: 0
rx_frame_errors: 0
hw_rsc_aggregated: 0
hw_rsc_flushed: 0
fdir_match: 911095
fdir_miss: 20
rx_fifo_errors: 0
rx_missed_errors: 0
tx_aborted_errors: 0
tx_carrier_errors: 0
tx_fifo_errors: 0
tx_heartbeat_errors: 0
tx_timeout_count: 0
tx_restart_queue: 460780
rx_long_length_errors: 0
rx_short_length_errors: 0
tx_flow_control_xon: 0
rx_flow_control_xon: 0
tx_flow_control_xoff: 0
rx_flow_control_xoff: 0
rx_csum_offload_errors: 0
alloc_rx_page_failed: 0
alloc_rx_buff_failed: 0
rx_no_dma_resources: 0
fcoe_bad_fccrc: 0
rx_fcoe_dropped: 0
rx_fcoe_packets: 0
rx_fcoe_dwords: 0
tx_fcoe_packets: 0
tx_fcoe_dwords: 0
tx_queue_0_packets: 2
tx_queue_0_bytes: 116
tx_queue_1_packets: 1
tx_queue_1_bytes: 42
tx_queue_2_packets: 645
tx_queue_2_bytes: 976258
tx_queue_3_packets: 48749596
tx_queue_3_bytes: 73792491352
tx_queue_4_packets: 0
tx_queue_4_bytes: 0
tx_queue_5_packets: 0
tx_queue_5_bytes: 0
tx_queue_6_packets: 5
tx_queue_6_bytes: 398
tx_queue_7_packets: 6
tx_queue_7_bytes: 660
rx_queue_0_packets: 2
rx_queue_0_bytes: 120
rx_queue_1_packets: 0
rx_queue_1_bytes: 0
rx_queue_2_packets: 0
rx_queue_2_bytes: 0
rx_queue_3_packets: 911110
rx_queue_3_bytes: 60150068
rx_queue_4_packets: 0
rx_queue_4_bytes: 0
rx_queue_5_packets: 0
rx_queue_5_bytes: 0
rx_queue_6_packets: 0
rx_queue_6_bytes: 0
rx_queue_7_packets: 5
rx_queue_7_bytes: 850
Receiver:
# ethtool -S p1p1
NIC statistics:
rx_packets: 48750250
tx_packets: 911123
rx_bytes: 73793468464
tx_bytes: 60151470
rx_pkts_nic: 48750250
tx_pkts_nic: 911123
rx_bytes_nic: 73988469464
tx_bytes_nic: 63795998
lsc_int: 2
tx_busy: 0
non_eop_descs: 0
rx_errors: 0
tx_errors: 0
rx_dropped: 0
tx_dropped: 0
multicast: 0
broadcast: 1
rx_no_buffer_count: 0
collisions: 0
rx_over_errors: 0
rx_crc_errors: 0
rx_frame_errors: 0
hw_rsc_aggregated: 0
hw_rsc_flushed: 0
fdir_match: 48750118
fdir_miss: 130
rx_fifo_errors: 0
rx_missed_errors: 0
tx_aborted_errors: 0
tx_carrier_errors: 0
tx_fifo_errors: 0
tx_heartbeat_errors: 0
tx_timeout_count: 0
tx_restart_queue: 2
rx_long_length_errors: 0
rx_short_length_errors: 0
tx_flow_control_xon: 0
rx_flow_control_xon: 0
tx_flow_control_xoff: 0
rx_flow_control_xoff: 0
rx_csum_offload_errors: 0
alloc_rx_page_failed: 0
alloc_rx_buff_failed: 0
rx_no_dma_resources: 0
fcoe_bad_fccrc: 0
rx_fcoe_dropped: 0
rx_fcoe_packets: 0
rx_fcoe_dwords: 0
tx_fcoe_packets: 0
tx_fcoe_dwords: 0
tx_queue_0_packets: 1
tx_queue_0_bytes: 42
tx_queue_1_packets: 907884
tx_queue_1_bytes: 59937360
tx_queue_2_packets: 0
tx_queue_2_bytes: 0
tx_queue_3_packets: 0
tx_queue_3_bytes: 0
tx_queue_4_packets: 0
tx_queue_4_bytes: 0
tx_queue_5_packets: 3232
tx_queue_5_bytes: 213600
tx_queue_6_packets: 6
tx_queue_6_bytes: 468
tx_queue_7_packets: 0
tx_queue_7_bytes: 0
rx_queue_0_packets: 2
rx_queue_0_bytes: 120
rx_queue_1_packets: 48564851
rx_queue_1_bytes: 73512807502
rx_queue_2_packets: 0
rx_queue_2_bytes: 0
rx_queue_3_packets: 0
rx_queue_3_bytes: 0
rx_queue_4_packets: 0
rx_queue_4_bytes: 0
rx_queue_5_packets: 185397
rx_queue_5_bytes: 280660842
rx_queue_6_packets: 0
rx_queue_6_bytes: 0
rx_queue_7_packets: 0
rx_queue_7_bytes: 0
I noticed that tx_restart_queue count seemed unusually high, so I
tried playing with the tx and rx ring size (both larger and smaller).
Increasing the tx ring size to 2048 eliminated the tx_restart_queue
issues but none of my changes had any effect on the reported
retransmission/reordering.
Thanks,
Jesse
------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
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
