performance / tuning tips. to the point.
|
Network Monitoring
Linux Windows AIX Solaris
Linux
Use 'netstat' to query network traffic on linux:
# netstat -i 2
Kernel Interface table
Iface MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR Flg
eth1 1500 0 6272524 0 0 0 68591 0 0 0 BMRU
lo 16436 0 12524 0 0 0 12524 0 0 0 LRU
eth1 1500 0 6272535 0 0 0 68601 0 0 0 BMRU
lo 16436 0 12524 0 0 0 12524 0 0 0 LRU
eth1 1500 0 6272538 0 0 0 68604 0 0 0 BMRU
lo 16436 0 12524 0 0 0 12524 0 0 0 LRU
|
WindowsWindows Task
Manager can be used to monitor your system. You can start by
"Start->Run->(type) taskmgr and hit <Enter>"
This screen shows two adapters. #3 has relatively heavy traffic (high at almost 17%).
|
To automate and capture network usage into a file using 'perfmon' tool that comes
with Windows 2003, see "perf automation" section from the left navigation.
AIX
The following
'netstat' example will report network activity for adapter en0 at a five
second interval for an unspecified duration.
# netstat -I en0 5
input (en1) output input (Total) output
packets errs packets errs colls packets errs packets errs colls
260984 0 128140 0 0 514023 0 405638 0 0
342 0 331 0 0 342 0 331 0 0
837 0 828 0 0 837 0 828 0 0
671 0 670 0 0 673 0 672 0 0
Where 'en0' is the network adapter you are interested in. There are usually more than one
network adapters in an AIX machine, for example, in an IBM pSeries p690.
|
Solaris
'netstat' can be used to determine if there maybe a network bottleneck
by looking at the 'total packets' column:
netstat –I hme0 5 5
input hme0 output input (Total) output
packets errs packets errs colls packets errs packets errs colls
11931 0 1470 0 405 15588 0 5127 0 405
3999 0 501 0 258 4009 0 511 0 258
4021 0 502 0 276 4031 0 512 0 276
4020 0 503 0 272 4030 0 513 0 272
4028 0 502 0 275 4038 0 512 0 275
|
|
|
|
