Using the netpmon command
The netpmon command will start tracing immediately unless the -d option is used.
Use the trcstop command to stop tracing. At that time, all the specified reports are generated, and the netpmon command exits. In the client-server environment, use the netpmon command to view how networking affects the overall performance. It can be run on both client and server.
The netpmon command can read the I/O trace data from a specified file, instead of from the real-time trace process. In this case, the netpmon report summarizes the network activity for the system and period represented by the trace file. This offline processing method is useful when it is necessary to postprocess a trace file from a remote machine or perform the trace data collection at one time and postprocess it at another time.
# gennames > gennames.out
# trcrpt -r trace.out > trace.rptAt this point, an adjusted trace logfile is fed into the netpmon command to report on I/O activity captured by a previously recorded trace session as follows:
# netpmon -i trace.rpt -n gennames.out | pgIn this example, the netpmon command reads file system trace events from the trace.rpt input file. Because the trace data is already captured on a file, the netpmon command does not put itself in the background to allow application programs to be run. After the entire file is read, a network activity report will be displayed on standard output (which, in this example, is piped to the pg command).
If the trace command was run with the -C all flag, then run the trcrpt command also with the -C all flag (see Formatting a report from trace -C output).
The following netpmon command running on an NFS server executes the sleep command and creates a report after 400 seconds. During the measured interval, a copy to an NFS-mounted file system /nfs_mnt is taking place.
# netpmon -o netpmon.out -O all; sleep 400; trcstopWith the -O option, you can specify the report type to be generated. Valid report type values are:
- cpu
- CPU usage
- dd
- Network device-driver I/O
- so
- Internet socket call I/O
- nfs
- NFS I/O
- all
- All reports are produced. The following is the default value.
# cat netpmon.out Fri Mar 5 15:41:52 2004 System: AIX crusade Node: 5 Machine: 000353534C00 ======================================================================== Process CPU Usage Statistics: ----------------------------- Network Process (top 20) PID CPU Time CPU % CPU % ---------------------------------------------------------- netpmon 45600 0.6995 1.023 0.000 nfsd 50090 0.5743 0.840 0.840 UNKNOWN 56912 0.1274 0.186 0.000 trcstop 28716 0.0048 0.007 0.000 gil 3870 0.0027 0.004 0.004 ksh 42186 0.0024 0.003 0.000 IBM.ServiceRMd 14966 0.0021 0.003 0.000 IBM.ERrmd 6610 0.0020 0.003 0.000 IBM.CSMAgentRMd 15222 0.0020 0.003 0.000 IBM.AuditRMd 12276 0.0020 0.003 0.000 syncd 4766 0.0020 0.003 0.000 sleep 28714 0.0017 0.002 0.000 swapper 0 0.0012 0.002 0.000 rpc.lockd 34942 0.0007 0.001 0.000 netpmon 28712 0.0006 0.001 0.000 trace 54622 0.0005 0.001 0.000 reaper 2580 0.0003 0.000 0.000 netm 3612 0.0002 0.000 0.000 aixmibd 4868 0.0001 0.000 0.000 xmgc 3354 0.0001 0.000 0.000 ---------------------------------------------------------- Total (all processes) 1.4267 2.087 0.844 Idle time 55.4400 81.108 ======================================================================== First Level Interrupt Handler CPU Usage Statistics: --------------------------------------------------- Network FLIH CPU Time CPU % CPU % ---------------------------------------------------------- external device 0.3821 0.559 0.559 PPC decrementer 0.0482 0.070 0.000 data page fault 0.0137 0.020 0.000 queued interrupt 0.0002 0.000 0.000 ---------------------------------------------------------- Total (all FLIHs) 0.4441 0.650 0.559 ======================================================================== Second Level Interrupt Handler CPU Usage Statistics: ---------------------------------------------------- Network SLIH CPU Time CPU % CPU % ---------------------------------------------------------- phxentdd32 2.4740 3.619 3.619 ---------------------------------------------------------- Total (all SLIHs) 2.4740 3.619 3.619 ======================================================================== Network Device-Driver Statistics (by Device): --------------------------------------------- ----------- Xmit ----------- -------- Recv --------- Device Pkts/s Bytes/s Util QLen Pkts/s Bytes/s Demux ------------------------------------------------------------------------------ ethernet 4 7237.33 10957295 0.0%27.303 3862.63 282624 0.2324 ======================================================================== Network Device-Driver Transmit Statistics (by Destination Host): ---------------------------------------------------------------- Host Pkts/s Bytes/s ---------------------------------------- client_machine 7237.33 10957295 ======================================================================== NFS Server Statistics (by Client): ---------------------------------- ------ Read ----- ----- Write ----- Other Client Calls/s Bytes/s Calls/s Bytes/s Calls/s ------------------------------------------------------------------------ client_machine 0.00 0 0.00 0 321.15 ------------------------------------------------------------------------ Total (all clients) 0.00 0 0.00 0 321.15 ======================================================================== Detailed Second Level Interrupt Handler CPU Usage Statistics: ------------------------------------------------------------- SLIH: phxentdd32 count: 33256 cpu time (msec): avg 0.074 min 0.018 max 288.374 sdev 1.581 COMBINED (All SLIHs) count: 33256 cpu time (msec): avg 0.074 min 0.018 max 288.374 sdev 1.581 ======================================================================== Detailed Network Device-Driver Statistics: ------------------------------------------ DEVICE: ethernet 4 recv packets: 33003 recv sizes (bytes): avg 73.2 min 60 max 618 sdev 43.8 recv times (msec): avg 0.000 min 0.000 max 0.005 sdev 0.000 demux times (msec): avg 0.060 min 0.004 max 288.360 sdev 1.587 xmit packets: 61837 xmit sizes (bytes): avg 1514.0 min 1349 max 1514 sdev 0.7 xmit times (msec): avg 3.773 min 2.026 max 293.112 sdev 8.947 ======================================================================== Detailed Network Device-Driver Transmit Statistics (by Host): ------------------------------------------------------------- HOST: client_machine (10.4.104.159) xmit packets: 61837 xmit sizes (bytes): avg 1514.0 min 1349 max 1514 sdev 0.7 xmit times (msec): avg 3.773 min 2.026 max 293.112 sdev 8.947 ======================================================================== Detailed NFS Server Statistics (by Client): ------------------------------------------- CLIENT: client_machine other calls: 2744 other times (msec): avg 0.192 min 0.075 max 0.311 sdev 0.025 COMBINED (All Clients) other calls: 2744 other times (msec): avg 0.192 min 0.075 max 0.311 sdev 0.025
The output of the netpmon command is composed of two different types of reports: global and detailed. The global reports list statistics as follows:
- Most active processes
- First-level interrupt handlers
- Second-level interrupt handlers
- Network device drivers
- Network device-driver transmits
- TCP socket calls
- NFS server or client statistics
The global reports are shown at the beginning of the netpmon output and are the occurrences during the measured interval. The detailed reports provide additional information for the global reports. By default, the reports are limited to the 20 most active statistics measured. All information in the reports is listed from top to bottom as most active to least active.