Output drops on a 3750-E 10G interface?

[RadiclDreamer]

Trying to find out why I am getting output drops on a 3750-E that is connected via 10G link back to a nexus 7010 via stub eigrp.

Show int on the nexus gives me the following, the reason for the int resets and the flaping was me changing optics to see if that was the culprit.

Let me know if you need the config etc of either int or any other info.

Ethernet1/19 is up
Hardware: 10000 Ethernet, address: 0026.51c7.3041 (bia 8843.e1b1.485a)
Description: Uplink to 3750-E
Internet Address is 172.25.253.109/30
MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA
full-duplex, 10 Gb/s, media type is 10g
Beacon is turned off
Auto-Negotiation is turned off
Input flow-control is off, output flow-control is off
Rate mode is shared
Switchport monitor is off
Last link flapped 00:04:03
Last clearing of "show interface" counters 31w4d
30 seconds input rate 195072 bits/sec, 96 packets/sec
30 seconds output rate 495360 bits/sec, 96 packets/sec
Load-Interval #2: 5 minute (300 seconds)
input rate 198.14 Kbps, 81 pps; output rate 1.43 Mbps, 96 pps
L3 in Switched:
ucast: 2312375651 pkts, 346627682876 bytes - mcast: 4143311 pkts, 349321356 bytes
L3 out Switched:
ucast: 1560143486 pkts, 1236940902253 bytes - mcast: 0 pkts, 0 bytes
RX
2312446750 unicast packets 5740269 multicast packets 7 broadcast packets
2318187026 input packets 347280331929 bytes
0 jumbo packets 0 storm suppression packets
0 runts 0 giants 0 CRC 0 no buffer
0 input error 0 short frame 0 overrun 0 underrun 0 ignored
0 watchdog 0 bad etype drop 0 bad proto drop 0 if down drop
0 input with dribble 0 input discard
0 Rx pause
TX
1560203052 unicast packets 5735485 multicast packets 17037 broadcast packets
1565955574 output packets 1237541819820 bytes
0 jumbo packets
0 output error 0 collision 0 deferred 0 late collision
0 lost carrier 0 no carrier 0 babble
0 Tx pause
3 interface resets









And for the int of the 3750-E

3750-E#sh int te1/0/1
TenGigabitEthernet1/0/1 is up, line protocol is up (connected)
Hardware is Ten Gigabit Ethernet, address is 0027.0df9.f1c1 (bia 0027.0df9.f1c1)
Description: Layer 3 uplink to NEXUS
Internet address is 172.25.253.110/30
MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive not set
Full-duplex, 10Gb/s, link type is auto, media type is 10GBase-SR
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:00, output 00:00:00, output hang never
Last clearing of "show interface" counters 00:04:20
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 124
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 913000 bits/sec, 150 packets/sec
5 minute output rate 183000 bits/sec, 117 packets/sec
37662 packets input, 24202827 bytes, 0 no buffer
Received 81 broadcasts (0 IP multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 81 multicast, 0 pause input
0 input packets with dribble condition detected
32898 packets output, 6561818 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 PAUSE output
0 output buffer failures, 0 output buffers swapped out
3750-E#

 

[RadiclDreamer]

Just a side note, on 12.2.(46)SE release notes I notice the changes list fixing the output drops displaying correctly as a fix, but as i am on 12.2.(53)SE2 I dont think this is a bogus issue, unless maybe the issue is still ongoing

 

[Railgun]

If it's that low of a rate consistently and you're not bursting at all, then I'd liken it to another quirk. If there's one thing I've learned, IOS is consistently inconsistent.

 

[spidey07]

Just a side note, on 12.2.(46)SE release notes I notice the changes list fixing the output drops displaying correctly as a fix, but as i am on 12.2.(53)SE2 I dont think this is a bogus issue, unless maybe the issue is still ongoing

I remember that bug. Probably best to just open a TAC case.

 

[RadiclDreamer]

I remember that bug. Probably best to just open a TAC case.

I was afraid I would hear that, management decided since the 3750-E had lifetime warranty and they were just closet switches to not get TAC on them, Its not causing any issues that im aware of, i just wanted to be proactive and get it resolved before it became a problem (if it even was one)

 

[spidey07]

I was afraid I would hear that, management decided since the 3750-E had lifetime warranty and they were just closet switches to not get TAC on them, Its not causing any issues that im aware of, i just wanted to be proactive and get it resolved before it became a problem (if it even was one)

Dumb, dumb, super dumb decision. It ALWAYS cost you more money in the long run.

 

[RadiclDreamer]

Dumb, dumb, super dumb decision. It ALWAYS cost you more money in the long run.

Well, not my decision to make, I had to drop that and the Wireless AP to get the stuff i really needed.

On the up side they did let me keep some spare 3750-E units and spare APS so i can swap out if necessary. Also, all we do is change vlans on the switches, never any config changes beyond that.

 

[Cooky]

Are you within your optical budget?
Since I don't see I/O errors on the interface, I don't think that's the issue, but do "show int transceiver" to confirm that.

Another possibility is your queueing.
3K's generally have queue 1 shaped to 4% by default.
So if you have traffic w/ CoS/DSCP values that are mapped to queue 1, they'd get dropped if volume reaches 4%.
Even though it's a 10G interface, 4% only gets you 400M.

 

[RadiclDreamer]

Got it figured out, seems to be a failing asic, try the other slot and it works fine, back to the original and it fails, thank John Chambers for free lifetime replacement on the 3750-E series....should be as much as they cost.