1 |
Paul Hartman posted on Thu, 06 Jun 2013 15:11:59 -0500 as excerpted: |
2 |
|
3 |
> On Thu, Jun 6, 2013 at 2:09 PM, Mark Knecht <markknecht@×××××.com> |
4 |
> wrote: |
5 |
>> Hi, |
6 |
>> Just taking a shot at the dark on this list before I ask something |
7 |
>> in the forums. Is there a simple app (or even something at the command |
8 |
>> line) that I can use to measure network throughput between two Gentoo |
9 |
>> machines on my internal network? |
10 |
> |
11 |
> Check out net-analyzer/ttcp and net-misc/iperf |
12 |
|
13 |
In addition to these which others mentioned, take a look at bing (NOT the |
14 |
MS search engine, "Bandwidth-PING"!). It's probably most useful outside |
15 |
the LAN once you've decided your LAN is fine, but it can be used inside |
16 |
as well, and bing can be /quite/ useful for looking at how latency is |
17 |
affected by packet size and/or content (compressible vs. not (pseudo- |
18 |
random), or you can specify the content). |
19 |
|
20 |
What's particularly nice on the WAN side is that you specify the near and |
21 |
the far end (neither of which you have to control), and bing tells you |
22 |
the difference in speed between them. So you can use a traceroute or the |
23 |
like to find the route taken, then focus in on segments of it. For |
24 |
instance, you can make the near end your ISP gateway and the far end the |
25 |
last ISP hop in your city, very useful for checking if there's a problem |
26 |
router in the local ISP's equipment. |
27 |
|
28 |
>> Background: We sold our house & moved. Comcast talked me into |
29 |
>> getting there new 'Blast' level Internet service with "speed up to |
30 |
>> 50Mb/S" but darned if it isn't slower than regular Comcast ISP service |
31 |
>> was a the previous house. In our house I typically got about 27Mb/S |
32 |
>> download using something like www.Speakeasy.net/speedtest at a |
33 |
>> measurement tool. Here I've never gotten higher than 22Mb/S. I do |
34 |
>> however get much better upload speeds - about 12Mb/S instead of the |
35 |
>> 5Mb/S I got at the house. |
36 |
> |
37 |
> I don't have Comcast but often ISPs will host a speed test server inside |
38 |
> their network, so you can ensure the speeds you're seeing are not being |
39 |
> limited by normal Internet slowdown issues outside of their system. |
40 |
|
41 |
FWIW as can probably be deduced from my mail address, I'm on cox, another |
42 |
cable ISP. Luckily for me (I happen to live in cox land, not comcast |
43 |
land), cox's internet service consistently comes out near the top in |
44 |
customer surveys, while comcast at least by reputation is rather nearer |
45 |
the bottom. So I've always felt fortunate that I'm in cox territory, not |
46 |
comcast's, tho I guess rather obviously not everyone's experience is so |
47 |
terrible with comcast or people would be finding other alternatives. |
48 |
|
49 |
> To take a page out of the generic ISP tech support, I would try plugging |
50 |
> your computer directly into the cable modem and seeing what kind of |
51 |
> speeds you get then, to eliminate any outside factors. |
52 |
|
53 |
Absolutely. This one was always pretty close to the first suggestion |
54 |
back when cox still had newsgroups and I hung out on them. |
55 |
|
56 |
> If you're using your own router, I would check to ensure it is fast |
57 |
> enough to handle that kind of speed. If it has Gigabit ethernet ports |
58 |
> that is usually a good sign. If it only has 10/100 then you might wind |
59 |
> up replacing it with something more modern. |
60 |
|
61 |
Strongly seconded once again. |
62 |
|
63 |
Because the router is normally doing NAPT (Network Address and Port |
64 |
Translation, aka PAT, Port Address Translation, the consumer level |
65 |
variant of the more generalized NAT, Network Address Translation) and |
66 |
often more active firewalling as well, and due to the cheap CPU and |
67 |
memory provisioning common consumer level routers have, very commonly the |
68 |
LAN/WAN thruput on a consumer level router is 50% or less the rated |
69 |
Ethernet port bandwidth. You'll often get near full port thruput on the |
70 |
LAN side as that's typically less router CPU processing, if any |
71 |
(sometimes the LAN side is simply an unmanaged switch, with the only real |
72 |
routing and processing actually done only between the LAN/WAN |
73 |
interfaces), but LAN/WAN thruput is all too commonly 25-33% port rating. |
74 |
|
75 |
Which means that a typical 100 Mb "fast ethernet" router will commonly |
76 |
top off at between 25-30 Mbit in real life. Thus, certainly Cox *VERY* |
77 |
strongly recommends a modern router with gigabit ports for all their |
78 |
higher tier internet services, as they're typically provisioned to do a |
79 |
couple hundred Mbit minimum (can't get /too/ far below the port rating) |
80 |
LAN/WAN thruput, while as I said it's VERY common to have 100 Mbit "fast |
81 |
ethernet" port routers top out at 25-30 Mbit. |
82 |
|
83 |
|
84 |
Meanwhile, I know a bit about cable modems from my cox newsgroup days as |
85 |
well. What brand and model modem do you have, and are you renting it or |
86 |
did you purchase? I'm personally quite partial to the Motorola Surf |
87 |
Board brand modems, as they tend to be reliably high quality, and to |
88 |
expose more troubleshooting information on the customer side if they know |
89 |
where to look. Other brand modems /can/ be as good, but the Motorola |
90 |
surf boards have seemed to have been consistently good quality (and as I |
91 |
said to expose more trouble shooting info to the customer) from the first |
92 |
dialup uplink cable modem models many years ago thru to the latest |
93 |
DOCSIS-3.x rated sb6xxx series. |
94 |
|
95 |
In your web browser, try going to http://192.168.100.1 (this assumes a |
96 |
stand-alone modem, a combined modem/router /may/ expose the same |
97 |
information differently, I'm not sure as I've never had one). On any |
98 |
DOCSIS certified modem, this should be the modem's internal web server |
99 |
troubleshooting interface (assuming comcast doesn't disable it entirely, |
100 |
cox doesn't). As I said above, however, some brand/model modems expose |
101 |
more information here than others, with the Motorola Surfboards being |
102 |
consistently really good. |
103 |
|
104 |
Depending on your modem's brand and model (and on what the ISP has |
105 |
configured as restricted), the interface will differ some. However, the |
106 |
most critical information is usually found on a signals page, or similar. |
107 |
|
108 |
There are three critical signal-strength numbers. In order of what tends |
109 |
to show problems first they are upstream power level, downstream power |
110 |
level, and downstream SNR (signal to noise ratio). |
111 |
|
112 |
Upstream power level is best in the 40s dBmV, tho I've seen people report |
113 |
connections at much lower values (into the lower 30s IIRC and even one at |
114 |
27, tho I wonder how he could connect at all or maybe his firmware was |
115 |
weird and it was mis-reporting, he was having issues, tho), and depending |
116 |
on the modulation used, the numbers typically top out at 55-58. Above 50 |
117 |
means your modem is effectively having to shout to be heard properly by |
118 |
the cable head-end, to the point it's causing interference with the |
119 |
downstream signal as well, while below about 38-40 means your modem is |
120 |
whispering and even that is still coming thru painfully loud at the other |
121 |
end. |
122 |
|
123 |
But an upstream power in the 40s is good. =:^) |
124 |
|
125 |
Downstream power is ideally 0, and the two ends will adjust their |
126 |
transmission power levels within a range to try to keep (near) zero at |
127 |
the other end if possible, so this one doesn't go out of range as often |
128 |
as upstream power, but if it does, it definitely indicates problems. The |
129 |
equipment is rated to work at zero +/- 15 dBmV, but from all I've seen, |
130 |
you want it between about -8 and +2 if possible -- if it gets out of that |
131 |
range you can usually still connect, but there tend to be more issues as |
132 |
the connection gets more marginal. A positive value isn't very common |
133 |
and often indicates an additional line-amp in the line -- line-amps are |
134 |
often useful for (at least old style analog, I don't really know about |
135 |
the newer digital, tho I suspect it may be more like internet) video, but |
136 |
tend to be more problematic for internet, thus the unbalance favoring the |
137 |
negative side. If you're better than -6, solid connection. -8, still |
138 |
pretty good but you might have occasional temporary issues. -10 is |
139 |
getting marginal and often means intermittent issues. -12 or worse, |
140 |
better be worried. |
141 |
|
142 |
Downstream SNR. Ultimately, this is the number that really counts, the |
143 |
number that the power dynamically adjusts for to keep consistent, tho of |
144 |
course you can only see the modem's side of this one, not the number at |
145 |
the head-end. Higher is better. Typical good numbers are in the upper |
146 |
30s, tho down to 32 or so should be usable. Honestly, I don't remember |
147 |
seeing this one go low too often, however, unless at least one of the |
148 |
other two were WAAYYY out, which isn't surprising, since by design the |
149 |
others adjust to try to keep this one in line, so the others will go out |
150 |
of line first. |
151 |
|
152 |
Finally, even if your numbers are reasonable, note whether they change |
153 |
dramatically over the course of hours. Some seasonal swing is normal -- |
154 |
colder typically better so summer is the critical time -- but if you're |
155 |
swinging 10 dBmV (or even 8, I'd actually be worried if it's more than 6) |
156 |
either upstream or downstream in a few hours and it's not due to some |
157 |
really serious weather changes, chances are good that there's a loose |
158 |
fitting or bad cable somewhere. The reason this matters even if the |
159 |
numbers stay reasonably good is that there's a limit to the dynamic |
160 |
adjustment the equipment can make while maintaining a connection. Thus, |
161 |
big swings often force the equipment to break the existing connection and |
162 |
renegotiate a new one with new parameters. That's fine if it's happening |
163 |
a time or two a season, but if it's happening several times a day, it's |
164 |
irritating, since you will obviously not be able to do anything on the |
165 |
net while it's renegotiating. |
166 |
|
167 |
So good numbers are 40s upstream power, 0 to -6 downstream power, and mid |
168 |
to upper 30s downstream SNR, without wild swings. |
169 |
|
170 |
With upstream power often being the first to show issues, if you're |
171 |
running over 50 or under 40 there, it's quite likely to be affecting your |
172 |
speeds. As I mentioned, even tho it's upstream, the two-way nature of |
173 |
the common TCP connection and the fact that there can be some |
174 |
interference between upstream and downstream does mean that an upstream |
175 |
power above 50, certainly above 52, can mean downstream issues as well. |
176 |
That has both my own experience and that I've seen on the cox newsgroups |
177 |
(and the general comp.dcom.xdsl and cable-modems newsgroups during the |
178 |
time I was reading them too) as well. |
179 |
|
180 |
Lastly, on Motorola Surfboards at least, and some but not all other |
181 |
brands, there's generally a log page available that can make interesting |
182 |
reading too. I won't cover it in nearly the detail I did the above but |
183 |
two hints for reading it: 1) At least Motorola Surfboard modems run Linux |
184 |
internally (modern versions even have an open source page with links to |
185 |
the appropriate sources, in compliance with the GPL... tho it doesn't do |
186 |
a lot of good unless you have a handy cable head-end lying around, since |
187 |
by DOCSIS standard, the firmware can only be flashed from the RFI side, |
188 |
not the Ethernet side), and thus have the same Linux/POSIX standard epoch |
189 |
time, January 1, 1970. Thus, any log events showing as 1970 indicate the |
190 |
modem wasn't able to contact a time server since its last reset at the |
191 |
time that event occurred, so it's effectively measuring the time since |
192 |
the modem booted, before it could get a connection. And AFAIK, times are |
193 |
in universal time, so offset from (basically) GMT, not local time. 2) |
194 |
Logged events can appear much more alarming than they actually are, and |
195 |
in fact, on some cable systems certain functions won't be used at all so |
196 |
will repeatedly timeout, unless/until the cableco decides to turn off the |
197 |
warnings entirely in the config, which it often does eventually, but not |
198 |
always right away. If you actually lose sync, either that or modem reset |
199 |
along with a few 1970 events until it can contact a time server again, |
200 |
show up. So if you're not seeing that very often (my log shows cox did |
201 |
something short early on June 6, but there weren't any 1970 times |
202 |
reported so it was short, and before that, the last outage was May 16, |
203 |
longer and more serious, as 1970 times reach back from then until the |
204 |
beginning of the log), nothing to worry about even if some of the logged |
205 |
events do look rather alarming. |
206 |
|
207 |
-- |
208 |
Duncan - List replies preferred. No HTML msgs. |
209 |
"Every nonfree program has a lord, a master -- |
210 |
and if you use the program, he is your master." Richard Stallman |