Very high "Drops TX" on a port on HP 2948 switch

[Loh Cheng Song]

Hi,

I have a HP Procurve 2948 switch running T.13.63.

A machine connected to switch port 30 stores data and it hand out chucks of data jobs to 26 PCs (connected to switch ports 1-26) to crunch. After processing the data, the 26 PC will write the data back to the machine connected to switch port 30.

With this setup, I keep getting the following alarms in the logs:

W 10/20/09 17:23:22 00331 FFI: port 30-High collision or drop rate. See help.
W 10/20/09 17:48:29 00331 FFI: port 30-High collision or drop rate. See help.
W 10/20/09 17:52:45 00331 FFI: port 30-High collision or drop rate. See help.


show interface:
show interfaces

Status and Counters - Port Counters

Flow Bcast
Port Total Bytes Total Frames Errors Rx Drops Tx Ctrl Limit
------- -------------- -------------- ------------ ------------ ----- ------
1 3,584,088,177 1,121,033,657 0 1662 off 0
2 3,175,408,209 952,848,813 0 1968 off 0
3 356,557,885 1,278,284,752 0 2222 off 0
4 1,371,912,853 1,182,642,999 0 1642 off 0
5 4,030,004,070 1,166,223,862 0 1847 off 0
6 3,691,949,309 1,408,614,550 0 1879 off 0
7 65,390,214 1,427,827,738 0 1393 off 0
8 1,714,705,759 900,902,421 0 2036 off 0
9 2,202,398,544 804,173,758 0 0 off 0
10 59,543,460 900,874,452 0 1802 off 0
11 4,159,462,783 1,113,716,478 0 924 off 0
12 3,164,934,515 1,270,553,348 0 1528 off 0
13 2,913,893,048 1,040,434,364 0 9806 off 0
14 2,251,493,163 1,363,271,604 0 11,577 off 0
15 1,806,898,577 1,441,208,953 0 2345 off 0
16 2,660,986,900 1,019,089,626 0 11,881 off 0
17 2,158,499,258 1,471,875,928 0 9953 off 0
18 3,090,298,334 1,456,183,233 0 9758 off 0
19 2,103,485,017 1,266,996,057 0 10,381 off 0
20 457,113,262 993,978,356 0 9485 off 0
21 3,520,976,498 1,021,109,721 0 10,471 off 0
22 3,217,735,840 1,261,790,775 0 8238 off 0
23 2,427,287,441 1,484,126,283 0 2910 off 0
24 1,164,169,552 1,149,173,153 0 2275 off 0
25 2,919,528,849 2,795,851,095 0 35,632 off 0
26 1,105,089,408 246,895,549 0 4843 off 0
27 1,499,137,488 150,153,341 0 0 off 0
28 3,208,725,425 811,046,567 0 0 off 0
29 2,720,578,395 120,211,259 0 0 off 0
30 2,122,374,221 523,117,071 0 857,489,109 on 0
31 114,090,331 980,056,692 0 1953 off 0
32 3,115,140,867 979,598,101 0 6637 off 0
33 3,581,710,015 1,047,462,358 0 1862 off 0
34 1,979,351,213 952,249,151 0 5379 off 0

show interface 30:

Status and Counters - Port Counters for port 30

Name :
MAC Address : 0016b9-0ccda2
Link Status : Up
Totals (Since boot or last clear) :
Bytes Rx : 1,167,250,242 Bytes Tx : 955,118,731
Unicast Rx : 3,792,361,033 Unicast Tx : 1,023,368,621
Bcast/Mcast Rx : 6193 Bcast/Mcast Tx : 2,348,438
Errors (Since boot or last clear) :
FCS Rx : 0 Drops Tx : 857,489,109
Alignment Rx : 0 Collisions Tx : 0
Runts Rx : 0 Late Colln Tx : 0
Giants Rx : 0 Excessive Colln : 0
Total Rx Errors : 0 Deferred Tx : 0
Others (Since boot or last clear) :
Discard Rx : 0 Out Queue Len : 0
Unknown Protos : 0
Rates (5 minute weighted average) :
Total Rx (bps) : 5152208 Total Tx (bps) : 6828920
Unicast Rx (Pkts/sec) : 47 Unicast Tx (Pkts/sec) : 124
B/Mcast Rx (Pkts/sec) : 0 B/Mcast Tx (Pkts/sec) : 1
Utilization Rx : 00.51 % Utilization Tx : 00.68 %

I have enabled flow-control on witch port 30 as well as setting qos depth to 2 queues but it didn't help:

show run:
module 1 type J86yyA
module 2 type J86xxA
module 3 type J90XXA
interface 30
flow-control
exit
...
fault-finder over-bandwidth sensitivity high
qos queue-config 2-queues

Is the high "Drops TX" of any concern here?

Thanks

[cenk sasmaztin]

Generally high collision or drop rate eror occur link sped and type disagreement and cable eror

what port 30 connect to it

please connect web browser interface your switch,double click this eror message
and read popup window