 |
» |
|
|
|
Enables or disables message tracing for the specified transport
entity. Syntax | |
NETCONTROL {TRACEON=type[,options]} [;{NI=niName [;PROT=niprot]} ]
{TRACEOFF } {NET=niName }
{GATE=gatehalfname }
{PROT=gprot } |
where the parameter option has the following options: [DISC][,[filename][,[recsize][,filesize] ]] |
Parameters | |
- TRACEON
Enables tracing for the one entity specified by
the NI, PROT, NET,
or GATE keywords, or for the control
process if none of those keywords are specified. The control process
will be started if it is not already running. This function cannot
be used to modify any parameters of tracing which has already been
enabled. If tracing is already enabled for the specified entity,
a "PREVIOUSLY ENABLED" error will occur. - type
(Required). Specifies the type of data to trace
from the specified entity. This field is made up of one or more
of the following key letters, concatenated, and entered in any order: M — Trace Messages
H — Trace Packet Header Data
D — Trace Packet Data
S — Trace State Transitions
B — Trace Buffers
N — Trace Nodal Management Events Recommended type setting is MHD. There is no default. - options
Specifies additional information about where to
put the collected trace data. There are several parameters. | | | |  | NOTE: A comma must precede a parameter
whenever (a) that parameter is included or (b) that parameter is
omitted but any other parameter which follows
it is included. | | | | |
- TRACEOFF
Disables previously enabled tracing for one entity,
which is specified by the NI, PROT, NET, or GATE keywords, or for
the control process if none of those keywords are specified. If
tracing is not enabled for the specified entity, a "NOT
TRACING" error will occur. - NI=niname
Specifies the name of a configured network interface
the trace will apply to. Enter any valid NI name from the NMMGR
Network Interface screen which is not a gateway half. If the specified
NI was not previously configured and started, a "NOT STARTED"
error will occur. Specifying NI=niname without
the ;PROT= option, or NET=niname,
starts tracing for the network interface itself. - NET=niName
Specifies the name of a configured network interface
which is not a gatehalf. Enter any valid NI name, as configured
with NMMGR. Using this parameter, the function applies only to the
network interface itself, not to any attached protocols. - GATE=gatehalfname
Specifies the name of a configured gateway half
network interface to start tracing on. Enter any valid gatehalf
NI name from the NMMGR Network Interface Configuration screen. If
the specified gatehalf NI was not previously configured and started,
a "NOT STARTED" error will occur. - PROT=gprot PROT=niProt
Specifies that a protocol is the pertinent entity
for each specified function to act on. Enter the name of the protocol,
as follows: - gprot
Specifies the name of one transport general protocol
to start tracing on. Valid inputs are TCP, UDP, PXP, or IPU. If
the specified protocol did not start or is not one of these inputs,
a "NOT ACTIVE" error will occur. - niprot
Specifies the name of one network interface protocol
to start tracing on; must be used in conjunction with the NI=niname
parameter. Valid inputs depend on the NI type, according to the
table below. If the specified protocol did not start, is not configured,
or is not one of these inputs, a "NOT ACTIVE"
error will occur. | NI Type: | Valid Network Interface Protocol Names |
|---|
| LAN | IP, PROBE, ARP | | TOKEN | IP, ARP | | FDDI | IP, ARP | | 100VG-AnyLAN | IP, PROBE, ARP | | 100Base-T | IP, PROBE, ARP | | ROUTER | IP, DIAL | | X.25 | IP, X25 | | GATEHALF | IP, DIAL | | LOOP | IP |
Discussion | |
The tracing functions allow you to enable collection of internal
information about what the various transport modules are doing,
or what packets are being sent and received at the transport level. Using TRACEON you instruct a specific module
not only to begin tracing, but also what kind of data to trace and
what file to put it in. Tracing continues until explicitly stopped
via a matching TRACEOFF command,
or until the specified module, or all of transport, is stopped.
If multiple modules had tracing enabled to capture a problem, stopping
transport is the usual way to stop all tracing. For most problems you will need to enable TCP tracing, and
for IP store-and-forward problems you should enable IP tracing;
see the examples for sample commands. For link-related problems
you should enable link tracing (see the LINKCONTROL
command). Other NS tracing can be enabled under the guidance of
your HP support representative. When tracing is enabled successfully, the name of the active
trace file is displayed. You should write this down as it will not
be repeated at TRACEOFF time; otherwise, to determine which trace
file contains the desired data, check trace file creation times
by using :LISTF NMTC####.PUB.SYS,3. As soon as your problem has been duplicated, you should stop
tracing to avoid having the file "wrap" and overwrite
the data. At completion of tracing, a trace file may be formatted
using the NMDUMP.PUB.SYS utility.
Much of the information traced will be meaningful only to HP support
personnel. Example | |
To enable TCP tracing, enter :NETCONTROL TRACEON=MHD; PROT=TCP
TRACE FILE IS NMTC0128.PUB.SYS. (NETXPORT 2000) |
To disable TCP tracing, enter :NETCONTROL TRACEOFF; PROT=TCP |
To enable control process tracing, TCP tracing, and IP tracing
on the "LAN1" NI, all to the same file, enter :NETCONTROL START; NET=LAN1
:NETCONTROL TRACEON=MHDSBN
TRACE FILE IS NMTC0129.PUB.SYS. (NETXPORT 2000)
:NETCONTROL TRACEON=MHD,DISC,NMTC0129.PUB.SYS; PROT=TCP
TRACE FILE IS NMTC0129.PUB.SYS. (NETXPORT 2000)
:NETCONTROL TRACEON=MHD,DISC,NMTC0129.PUB.SYS; NI=LAN1; PROT=IP
TRACE FILE IS NMTC0129.PUB.SYS. (NETXPORT 2000) |
To disable all this tracing once enabled, enter :NETCONTROL TRACEOFF; NI=LAN1; PROT=IP
:NETCONTROL TRACEOFF; PROT=TCP
:NETCONTROL TRACEOFF |
|
