This page is of historical interest only. I stopped using DSL line when I left Wantadilla in July 2007.

In June 2005, after years of waiting, I finally got an ADSL connection to my home. Contrary to my fears, the line quality was more than ample. The Linksys AG241 gateway that I got has a web page showing the link status. Initially it showed an “Upstream Margin” of 23 dB and a “Downstream Margin” of 11 dB. They don't document what they mean by this term, but my guess is that it is the difference between the line attenuation and some arbitrary value at which they decide no signal is available; this would then be 73 dB downstream and 46 dB upstream. If anybody can clarify, I would be grateful.

The link worked flawlessly from the first time I plugged it in, though the technical and commercial problems with routing a network down it make interesting reading.

On 21 June, exactly two weeks after installation, that changed: the noise levels increased, the noise margins decreased accordingly, and the upstream speed dropped from an already low 256 kb/s to 160 kb/s. Coupled with the weather (heavy rain), all this looked like the scenario we have seen so many times in the past: telephone lines with damaged insulation get wet and fail. In the past, Telstra has been less than satisfactory in resolving this problem; in particular, they almost never uphold their Customer Service Guarantee, which requires faults to be repaired in two business days. The DSL gateway that I use provides line quality statistics, so this page helps document the current state. The graphs show:

• Upstream and downstream noise margins. These are an indication of the condition of the physical line. For reasons I haven't understood, but which probably have something to do with the router itself, the downstream margin fluctuates a lot, whereas the upstream rate is relatively steady.
• The upstream and downstream data rates. These should remain constant unless the line gets really bad.
• Link status. This is the connectivity to the other end of the DSL link, measured by an ICMP echo (ping) packet once a minute. This should remain steady at 1. A drop to 0 indicates a link failure.
• Network status. This is the number of remote systems responding to a ICMP echo (ping) packet at the same time as the link status ping. I currently ping 5 systems which I frequently access: freefall.FreeBSD.org, shell.mysql.com, www.auug.org.au, ozlabs.org and ftp.NetBSD.org. Normally this value should be 5. If it's less, the script retries every second until full connectivity is restored.

Click on the graphs for a 1600x1200 version.

The monitor scripts

The scripts that do this monitoring are:

• The main script pulls down the appropriate web page from the DSL router and runs gnuplot to create the graphs.
• The script linkcheck performs the periodic pings to check link and network status.
• The main script takes a raw gnuplot scripts and modifies it to create a total of 16 plots: four times (4 hours, 24 hours, one week and forever), two views (smoothed and unsmoothed) and two sizes (540x360 and 1600x1200). I chose 540x360 to fit two plots comfortably side-by-side on my screen; YMMV.