loadnet prints out load averages that describe the current saturation of the network interfaces. The value reported is based on queue errors due to saturation. The first example shows loadnet running with a sample interval of 1 second on a quiet system. There is no history of samples in the kernel, so loadnet needs to run for some time (15 minutes) before it can fill all the columns. Here I had several ssh sessions active, and was browsing the Internet. We also run netstat for comparison. $ ./loadnet 1 Net Time 1sec 5sec 15sec 1min 5min 15min 00:46:57 0.00 00:46:58 0.00 00:46:59 0.00 00:47:00 0.00 00:47:01 0.00 0.00 00:47:02 0.00 0.00 00:47:03 0.00 0.00 00:47:04 0.00 0.00 00:47:05 0.00 0.00 00:47:06 0.00 0.00 00:47:07 0.00 0.00 00:47:08 0.00 0.00 00:47:09 0.00 0.00 00:47:10 0.00 0.00 00:47:11 0.00 0.00 0.00 00:47:12 0.00 0.00 0.00 00:47:13 0.00 0.00 0.00 00:47:14 0.00 0.00 0.00 ^C $ netstat -iI hme0 1 input hme0 output input (Total) output packets errs packets errs colls packets errs packets errs colls 9332310 1 8047343 0 0 23043316 1 21758349 0 0 4 0 5 0 0 10 0 11 0 0 2 0 4 0 0 8 0 10 0 0 3 0 4 0 0 9 0 10 0 0 2 0 6 0 0 9 0 13 0 0 2 0 4 0 0 6 0 8 0 0 2 0 4 0 0 7 0 9 0 0 2 0 4 0 0 8 0 10 0 0 4 0 4 0 0 10 0 10 0 0 7 0 9 0 0 13 0 15 0 0 95 0 97 0 0 103 0 105 0 0 24 0 32 0 0 30 0 38 0 0 14 0 18 0 0 20 0 24 0 0 2 0 4 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 ^C loadnet has shown that the network interfaces were never saturated during this sample, netstat has shown that there was a little activity. In this example we run loadnet with a 2 second interval. As loadnet runs an 8Mb file was transferred using FTP. This generated load as the application trys to send data as fast as possible. netstat was also run for comparison. $ ./loadnet 2 Net Time 1sec 5sec 15sec 1min 5min 15min 00:56:29 0.00 00:56:31 0.00 00:56:33 0.00 0.00 00:56:35 0.00 0.00 00:56:37 0.00 0.00 00:56:39 0.00 0.00 00:56:41 0.00 0.00 00:56:43 0.00 0.00 0.00 00:56:45 1.64 0.40 0.13 00:56:47 1.64 1.06 0.35 00:56:49 1.60 1.65 0.58 00:56:51 1.55 1.64 0.79 00:56:53 0.00 1.17 0.86 00:56:55 0.00 0.52 0.86 00:56:57 0.00 0.00 0.86 00:56:59 0.00 0.00 0.84 00:57:01 0.00 0.00 0.61 00:57:03 0.00 0.00 0.39 ^C $ netstat -iI hme0 2 input hme0 output input (Total) output packets errs packets errs colls packets errs packets errs colls 9336458 1 8057136 0 0 23050925 1 21771603 0 0 6 0 7 0 0 18 0 19 0 0 24 0 28 0 0 38 0 42 0 0 22 0 26 0 0 34 0 38 0 0 10 0 12 0 0 22 0 24 0 0 24 0 27 0 0 36 0 39 0 0 30 0 33 0 0 42 0 45 0 0 186 0 1559 0 0 198 0 1571 0 0 184 0 1592 0 0 196 0 1604 0 0 184 0 1592 0 0 196 0 1604 0 0 173 0 1469 0 0 185 0 1481 0 0 7 0 10 0 0 19 0 22 0 0 2 0 4 0 0 14 0 16 0 0 2 0 4 0 0 14 0 16 0 0 ^C In this example we run loadnet and netstat with 1 second intervals. Then the spray command is used to send a quick flood of network packets to another server. This has generated a spike in the one second averages of 6.20, which has reduced to 0.41 by the 15 second averages. $ ./loadnet 1 Net Time 1sec 5sec 15sec 1min 5min 15min 01:01:21 0.00 01:01:22 0.00 01:01:23 0.00 01:01:24 0.00 01:01:25 0.00 0.00 01:01:26 0.00 0.00 01:01:27 6.20 1.24 01:01:28 0.00 1.24 01:01:29 0.00 1.24 01:01:30 0.00 1.24 01:01:31 0.00 1.24 01:01:32 0.00 0.00 01:01:33 0.00 0.00 01:01:34 0.00 0.00 01:01:35 0.00 0.00 0.41 01:01:36 0.00 0.00 0.41 01:01:37 0.00 0.00 0.41 ^C $ netstat -iI hme0 1 input hme0 output input (Total) output packets errs packets errs colls packets errs packets errs colls 9339509 1 8071540 0 0 23055728 1 21787759 0 0 4 0 5 0 0 10 0 11 0 0 4 0 7 0 0 10 0 13 0 0 2 0 4 0 0 8 0 10 0 0 7 0 404 0 0 13 0 410 0 0 6 0 776 0 0 11 0 781 0 0 2 0 4 0 0 6 0 8 0 0 3 0 5 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 2 0 6 0 0 10 0 14 0 0 2 0 4 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 2 0 4 0 0 8 0 10 0 0 ^C