Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials
Store

Oroin Showing Discards on 6500 Series Switches

Just wandering how many people here have any 6500 Series switch that are show discards?

We are currently running 8 6509 and 4 6513's and I am seeing a lot of discards on the interface ports, but when I look at the interface on the switch I see it running clean no errors, giants, runts etc....

I am currently trying to figure out how to get rid of these errors and I am at a stand still, as I am not sure where to even start looking to try to troubleshoot this and get rid of the errors.

These switchs are running IOS and 2 are running CATOS and I see errors on both types, the current queueing on these switches are FIFO there is no QOS setup on these switches.

Some of the ports that are showing errors are high utilized and some are not <5%.

Any help on this would be great.

Find more posts tagged with

Accepted answers

All comments

aLTeReGo

I'm sure the errors are real. The command "show counters slot/interface " will show you all the types of errors, and how many you have on an interface. Give it a shot. What is the error type being reported by CatOS?

Example

CISCO-6509# (enable) sh counters 3/16

64 bit counters

0 rxHCTotalPkts = 2

1 txHCTotalPkts = 64

2 rxHCUnicastPkts = 0

3 txHCUnicastPkts = 3

4 rxHCMulticastPkts = 0

5 txHCMulticastPkts = 59

6 rxHCBroadcastPkts = 0

7 txHCBroadcastPkts = 2

8 rxHCOctets = 136

9 txHCOctets = 4891

10 rxTxHCPkts64Octets = 1

11 rxTxHCPkts65to127Octets = 62

12 rxTxHCPkts128to255Octets = 1

13 rxTxHCPkts256to511Octets = 2

14 rxTxHCpkts512to1023Octets = 0

15 rxTxHCpkts1024to1518Octets = 0

16 txHCTrunkFrames = 34

17 rxHCTrunkFrames = 2

18 rxHCDropEvents = 0

32 bit counters

0 rxCRCAlignErrors = 0

1 rxUndersizedPkts = 0

2 rxOversizedPkts = 0

3 rxFragmentPkts = 0

4 rxJabbers = 0

5 txCollisions = 0

6 ifInErrors = 0

7 ifOutErrors = 0

8 ifInDiscards = 0

9 ifInUnknownProtos = 0

10 ifOutDiscards = 0

11 txDelayExceededDiscards = 0

12 txCRC = 0

13 linkChange = 0

14 wrongEncapFrames = 0

0 dot3StatsAlignmentErrors = 0

1 dot3StatsFCSErrors = 0

2 dot3StatsSingleColFrames = 0

3 dot3StatsMultiColFrames = 0

4 dot3StatsSQETestErrors = 0

5 dot3StatsDeferredTransmisions = 0

6 dot3StatsLateCollisions = 0

7 dot3StatsExcessiveCollisions = 0

8 dot3StatsInternalMacTransmitErrors = 0

9 dot3StatsCarrierSenseErrors = 0

10 dot3StatsFrameTooLongs = 0

11 dot3StatsInternalMacReceiveErrors = 0

12 dot3StatsSymbolErrors = 0

0 txPause = 0

1 rxPause = 0

0 rxTotalDrops = 2

1 rxFIFOFull = 0

2 rxBadCode = 0

Last-Time-Cleared

--------------------------

Mon Mar 13 2006, 19:31:23

Try posting your counters for the interface like I did above

Malvado

Ok I can see the errors, but I guess now I need to figure out how to the discards to stop.

1.                          InPackets = 918280819
2.                           InOctets = 216287035605
3.                        InUcastPkts = 915606638
4.                        InMcastPkts = 2647356
5.                        InBcastPkts = 26825
6.                         OutPackets = 1637383737
7.                          OutOctets = 917351338688
8.                       OutUcastPkts = 1637379548
9.                       OutMcastPkts = 0
10.                       OutBcastPkts = 4189
11.                           AlignErr = 0
12.                             FCSErr = 0
13.                            XmitErr = 0
14.                             RcvErr = 0
15.                          UnderSize = 0
16.                          SingleCol = 0
17.                           MultiCol = 0
18.                            LateCol = 0
19.                       ExcessiveCol = 0
20.                       CarrierSense = 0
21.                              Runts = 0
22.                             Giants = 0
23.                         InDiscards = 0
24.                        OutDiscards = 8348975

I guess FIFO is no longer going to cut it for a queueing method.

aLTeReGo

Is that a full list of the interface counters? It looks like you're missing some counters that might help troubleshoot your problem.

jtimes

Malvado,

Are the interfaces you are seeing these discards on connected to other switches, routers, servers, and or Firewalls?

Transmit Discards - indicate that packets were not transmitted because of network congestion. It's not the switch, it's the receiving station that can't keep up. We usually see this condition on servers that do not have up to date or incompatible NIC card drivers and or outdated Firmware. If the receiving station is a router you might want to investigate increasing the Input Hold-Queue and Output Hold-Queue for that interface on the router. The same applies for switch to switch connections in/out Hold Queues...

From Cisco:

"In the Catalyst 6500 architecture, access into the switch fabric itself is almost never the bottleneck. Rather, on the transmit side, one or several ports are the likely destination for a majority of the packets entering the switch. As such, the receive-side port buffers on the Ethernet modules are relatively small compared to the transmit-side port buffers."

willhepptemp

It's not the switch, it's the receiving station that can't keep up.

I have to disagree here. Out discards happen when the egress bandwidth + egress buffer space is completely utilized (i.e., the switch simply has no capacity to transmit a well-formed frame so has to "discard" it.) The receiving station does not 'cause' the problem - the swtich has no knowledge of the buffer capacity or other resources of the reciever - the receiver is simply oversubscribed. This can happen to both hosts as well as uplinks.

On a related note, discards are specifically not errors. Errors imply malformed frames (runts, giants, CFC, etc) whereas discards are well formed frames that simply cannot be forwarded.

Will

jpratt

We found that we had high discards on interfaces connected to firewalls, routers and other switches using this module: WS-X6148-RJ-45. When we move the interfaces to our WS-X6148-GE-TX modules the problem went away. The problem is caused by the buffers on the non gig modules are small and are exhausted quickly so bytes are discarded.

jtimes

agree to disagree... restating what i stated is not a disagreement.

egress etymology: Latin egressus, from egredi to go out, from e- + gradi to go

egress = output

Knowledge Base:

http://support.solarwinds.net/kbase/ProblemDetail.cfm?ID=406

salavie - as in life goes on...

aLTeReGo

I had a similar issue when I connected several servers to a WS-X6548-GE-TX blade. The WS-X6548-GE-TX is designed as a workstation/desktop switch module. The WS-X6548-GE-TX can handle a maximum of 6 Gbps to the back plane. There is an 8:1 over subscription as the ports are connected at gig rate, but only can supply 6 Gbps to the back plane at any given time. The buffers on this blade were also very small for server connectivity. 1MB of shared buffer memory for each group of 8 ports which contributed significantly to the problem.

I purchased some WS-X6748-GE-TX blades which have 1.3MB Buffer Memory per Port, and migrated the servers to them and haven't had any problem since. Hope this helps

Malvado

I see this issue on server ports rather than uplink ports, all my firewall and trunks, etc.. are clean as a whistle.

I am going to be moving one of my servers that is having an issue onto a gig module and see if it goes away, I also have different hardware versions of the same line cards. I will be moving this server around to different modules to see if it is one hardware version that is causing the issue or not.

I will update shortly.

mezdem

Only just started using Orion.. have had one port on a switch that needed upgrading to gib from 100mb due to transmit discards. this is resolved, but a new issue has appeared on one or our cores...

Our 6509 CatOS switch is now reporting 000's discards on Port 2/8 (module WS-X6408A-GBIC:3.0), its a 1gb link to our 2nd site..The line is approx 30-40% TX utilization ..

Dist_6509-A (enable) sh counters 2/8
64 bit counters
0 rxHCTotalPkts                      =         160738677837
1 txHCTotalPkts                      =         185441962402
2 rxHCUnicastPkts                    =         160419513166
3 txHCUnicastPkts                    =         184799378934
4 rxHCMulticastPkts                  =            221313394
5 txHCMulticastPkts                  =            434242713
6 rxHCBroadcastPkts                  =             97851183
7 txHCBroadcastPkts                  =            208340755
8 rxHCOctets                         =      128781295168512
9 txHCOctets                         =      200158973658683
10 rxTxHCPkts64Octets                 =               476454
11 rxTxHCPkts65to127Octets            =         100680535672
12 rxTxHCPkts128to255Octets           =          18203091445
13 rxTxHCPkts256to511Octets           =           1522157706
14 rxTxHCpkts512to1023Octets          =          31358552101
15 rxTxHCpkts1024to1518Octets         =           4427396163
16 txHCTrunkFrames                    =         185441724284
17 rxHCTrunkFrames                    =         160738439511
18 rxHCDropEvents                     =                    0
32 bit counters
0 rxCRCAlignErrors                   =          0
1 rxUndersizedPkts                   =          0
2 rxOversizedPkts                    =          0
3 rxFragmentPkts                     =          0
4 rxJabbers                          =          0
5 txCollisions                       =          0
6 ifInErrors                         =          0
7 ifOutErrors                        =          0
8 ifInDiscards                       =          0
9 ifInUnknownProtos                  =          0
10 ifOutDiscards                      =   34915720
11 txDelayExceededDiscards            =          0
12 txCRC                              =          0
13 linkChange                         =          0
14 wrongEncapFrames                   =          0
0 dot3StatsAlignmentErrors           =          0
1 dot3StatsFCSErrors                 =          0
2 dot3StatsSingleColFrames           =          0
3 dot3StatsMultiColFrames            =          0
4 dot3StatsSQETestErrors             =          0
5 dot3StatsDeferredTransmisions      =          0
6 dot3StatsLateCollisions            =          0
7 dot3StatsExcessiveCollisions       =          0
8 dot3StatsInternalMacTransmitErrors =          0
9 dot3StatsCarrierSenseErrors        =          0
10 dot3StatsFrameTooLongs             =          0
11 dot3StatsInternalMacReceiveErrors =          0
12 dot3StatsSymbolErrors              =          0
0 txPause                            =          0
1 rxPause                            =          0
0 rxTotalDrops                       =        107
1 rxFIFOFull                         =          0
2 rxBadCode                          =          0