Evaluating Network Performance

Some of the more common support questions we field concern network performance. Network performance issues have different symptoms on 10BaseT networks cards than 100BaseT.

If your system is utilizing a 10BaseT card you want verify that the transceiver has SQE Heartbeat enabled. If SQE Heartbeat is not present your system will suffer from poor network throughput. Most transceivers have DIP switch settings to control SQE functionality. On an HP Ethertwist transceiver you should confirm that SQE Test and Link Beat are both enabled. You can check the status of SQE Heartbeat using the :LINKCONTROL command. The count for Heartbeat losses should be zero.


Wait for about a minute then continue. This RESET parameter will initialize all of the statistics to zeroes so that all of the subsequent analysis can be performed on current data.


Linkname: DTSLINK   Linktype: IEEE8023  Linkstate: CONNECTED
Physical Path:              56/56
Current Station Address:    08-00-09-87-2B-F6
Default Station Address:    08-00-09-87-2B-F6
Current Receive Filter:     broad(1) any(0) k_pckts(1) x_pckts(0)
Current Multicast Addresses:
 09-00-09-00-00-01  09-00-09-00-00-03  09-00-09-00-00-04
Transmits no error       184658  Receives no error       614571
Transmit byte count    20668184  Receive byte count   868768653
Transmits error               0  Receives error               0
Transmits deferred          459  Carrier losses               0
Transmits 1 retry           845  CRC errors                   0
Transmits 1 retry           524  Frame losses                 0
Trans 16 collisions           0  Whole byte errors            0
Trans late collision          0  Size range errors            0
802 chip restarts             0  Receives dropped             0
Heartbeat losses              0  Receives broadcast       18346
                                 Receives multicast          55

Aside from Hearbeat losses, other statistics to keep an eye on are CRC errors, Transmits deferred, and all of the ‘error’ fields. A complete description for each of these statistics can be found in the NS 3000/iX Operations & Maintenance Reference Manual under the LINCONTROL command. Or you can view it online at (http://docs.hp.com/mpeix/onlinedocs/36922-90039/00/00/61-con.html.)

Effects of SQE on Data Terminal Controllers

Lack of SQE Heartbeat on DTCs can cause system performance problems as well. The DTC will ‘complain’ to the host system that it is missing SQE. The host system, your HP3000, will log the heartbeat loss events to special log files stored on LDEV 1. These log events occur continuously resulting in an I/O bottleneck on the system disk. Frequently the process that is logging the errors appears as the top DISC consumer in SOS or Glance/iX. Or a system process will continually show up in a list of active processes as seen in the :SHOWQ command.


DORMANT                                   RUNNING

Q  PIN   JOBNUM                           Q  PIN   JOBNUM

                                          A   39
                                          C  M163  #S9136
                                          C  M183  #S9140
                                          D  U189  #J6036

One last method to investigate if you have this problem is to check for the log files themselves. The system will one set of log file for each DTC configured on the system. The names of the log files are HxxxxxxA.PUB.SYS and HxxxxxxB.PUB.SYS where ‘xxxxxx’ represents the last 6 characters of the 12-digit Ethernet/MAC address of the DTC. For instance, if the MAC address 08-00-09-00-75-BD then the file name will be H0075BDA.PUB.SYS. If the EOF of this file is very large then you should verify the SQE settings on the transceiver connected to that DTC.

:listf h@a.pub.sys,2

FILENAME  CODE  ------------LOGICAL RECORD-----------  ----SPACE----
                  SIZE  TYP        EOF      LIMIT R/B  SECTORS #X MX

H0075BDA*           1W  FB           5      66010   1      256  1  *

If you would like to know more about SQE Heartbeat you can read about it on Cisco Systems web site at http://www.cisco.com/warp/public/86/12.html.

100Base-T Cards

100BaseT networks are not prone to Heartbeat Loss problems. But they are subject to packet collisions produced by conflicting configurations. 100BaseT cards can run at either 10Mbs or 100Mbs. Furthermore, at 100Mbs they run at either half- or full-duplex. Both sides of the conversation using opposing protocols introduces performance degradation. In fact, the performance of a 100BaseT card will have throughput less than 10Mps in a conflicting configuration. Typically this problem appears when the components are set to auto-negotiate the speed and duplex. Auto negotiation is simply not reliable.

You can use the :LINKCONTROL command to view the current status of you 100BaseT card:

<<wait about a minute>>

Linkname: BT100     Linktype: 100BT     Linkstate: CONNECTED
Physical Path:              10/4/4
Current Station Address:    00-10-83-2C-E6-32
Default Station Address:    00-10-83-2C-E6-32
Current Multicast Addresses:

Transmit bytes               24182167    Receive bytes                42617211
Transmits                      621343    Receives                       622469
Transmits no error             602358    Receives broadcast               1860
Transmits dropped                   0    Receives multicast                  0
Transmits deferred                  0    Receives no error              246570
Transmits 1 retry                   0    CRC or Maxsize error                0
Transmits 1 retry                   0    Code or Align error                 0
Trans 16 collisions                 0    Recv dropped: addr               1925
Trans late collision                0    Recv dropped: buffer                0
Trans underruns                     0    Recv dropped: dma                   0
Carrier losses                      0    Recv dropped: other                 4
Link disconnects                    0    Recv deferred                       0
Link speed                        100    Recv overruns                       0
Link duplex                      Half    Link auto sensed                  Yes
Link mode                  100Base-TX    Secs since clear                 8289

The correction for this problem is to disable auto-negotiation for the network card configuration in NMMGR, forcing 100Mbs Full-Duplex mode. The same adjustment is made to the hub or switch that the system is connected to. Go to the LINK configuration screen in NMMGR and check/modify the following highlighted fields:

Path:  LINK.BT100

Physical path of device adapter [32            ]

Use factory-configured local station address? [Y] (Y/N)
Local station address  [FF-FF-FF-FF-FF-FF] (Hex)

When auto-negotiation is enabled, the system can only
properly configure the link if the hub also auto-negotiates.
Use auto-negotiation to determine link settings? [N]  (Y/N)
If 'N':  Link speed   [100 ]   (100 or 10 MBits/sec)
         Full Duplex mode [Y]  (Y/N; N=Half)

Trace at startup?  [N] (Y/N)                       Note : Trace reduces
Trace filename     [                          ]           performance.


You must validate the NMCONFIG file (both DTS and NETXPORT) after making changes to LAN adapters. You also need to restart the network to activate the changes.

If you need help diagnosing if your system is victim to any of these network issues, or you would like to discuss any other network issues you may be experiencing please don’t hesitate to call upon us.