News items for tag network - Koos van den Hout

2013-05-08 (#) 7 years ago
Monoskop
multicast stream test image: Philips PM5544 testcard I had to test some part of the multicast setup on the work network and fired up VLC.

And from the department of 'some things never change' or 'still going strong', Bratislava STV2 testcard is still available under the title 'Monoskop' on vlc url rtp://@233.10.47.73:1234. This stream must have been going for years, I remember using it in tests years ago.

And during the day when STV2 has no program, it still transmits an actual Philips PM5544 testcard.

Tags: , , ,
2013-03-09 Searching for source of interference and PLC network 8 years ago
In the search for the source of the interference I do note the Devolo dlan powerline network has lousy performance when I compare the speed the devolo utilities say I will get compared to what iperf says:
koos@metcalfe:~$ dlanlist eth0
Type    MAC address        Mbps TX/RX       Version/Product
local   00:0B:3B:5F:95:AB  ---.-- / ---.--  INT6000-MAC-3-3-3348-00-2764-20080808-FINAL-B devolo dLAN 200 AVplus [MT2165]
remote  00:0B:3B:6F:AE:90   73.50 / 112.88  INT6000-MAC-3-3-3348-00-2764-20080808-FINAL-B
And from iperf:
[  3]  0.0-10.2 sec  1.43 MBytes  1.18 Mbits/sec
Interesting difference of opinion there, 73/112 megabit versus 1.18 megabit.

Update: it helps when I remember the setup of the system: I installed the wondershaper on that system which throttles bandwidth at .. 1 megabit. Re-running the test with the throttle disabled gives totally different results:
[  3]  0.0-120.1 sec    308 MBytes  21.5 Mbits/sec
Testing with udp gives even higher speeds:
[  3]  0.0-120.0 sec  1.34 GBytes  95.6 Mbits/sec
And in the other direction:
[  3]  0.0-120.3 sec    649 MBytes  45.2 Mbits/sec  0.204 ms 1075187/1538185 (70%)

Reception of 2 meter radio doesn't get worse/better during these tests. According to the devolo dlan 200 faq the devices work in the range of 0 to 30 MHz with notches for HF amateur frequencies. With the software defined radio I also see no change in the 2 meter band when running bandwidth tests. Which doesn't say a lot: I can't even find the output from PI3UTR at 145.625 MHz in the output plot, even with a receiver nearby receiving that same signal fine. The output plot does show APRS active on 144.800 MHz.

Tags: , ,
2012-12-18 (#) 8 years ago
Another thing I puppetized: setting static routes on redhat-like servers. We have two rfc1918 ranges for management network so systems with a public IPv4 address and a management IPv4 address need a static route to the other half of the management network. Time for puppet which distributes the /etc/sysconfig/network-scripts/route-* files and does an ifdown and ifup of the affected interface. Using the variables from facter I can find whether a host is on one of the management IPv4 ranges and on which interface, and create routes accordingly.

Tags: , , ,
2012-11-10 (#) 8 years ago
I just tried to visit Laser charged glowing display - Hack a Day but the page kept loading forever. The problem is with loading page content (images and scripts) from s0.wp.com, s1.wp.com, s2.wp.com. All of which seem to be part of wordpress hosting and come via the edgecast content distribution network:
koos@machiavelli:~$ host s0.wp.com
s0.wp.com is an alias for cs82.wac.edgecastcdn.net.
cs82.wac.edgecastcdn.net has address 93.184.220.111
cs82.wac.edgecastcdn.net has IPv6 address 2606:2800:234:1922:15a7:17bf:bb7:f09
koos@machiavelli:~$ host s1.wp.com
s1.wp.com is an alias for cs82.wac.edgecastcdn.net.
cs82.wac.edgecastcdn.net has address 93.184.220.111
cs82.wac.edgecastcdn.net has IPv6 address 2606:2800:234:1922:15a7:17bf:bb7:f09
koos@machiavelli:~$ host s2.wp.com
s2.wp.com is an alias for cs82.wac.edgecastcdn.net.
cs82.wac.edgecastcdn.net has address 93.184.220.111
cs82.wac.edgecastcdn.net has IPv6 address 2606:2800:234:1922:15a7:17bf:bb7:f09
And all suffer from a reachability problem via IPv6:
koos@machiavelli:~$ telnet s0.wp.com 80
Trying 2606:2800:234:1922:15a7:17bf:bb7:f09...
Taking a while to fall back to IPv4, which results in long page loading times. I noticed this on other sites too, like pages using gravatar. Which uses the same cdn with the same problem:
koos@machiavelli:~$ host 0.gravatar.com
0.gravatar.com is an alias for cs91.wac.edgecastcdn.net.
cs91.wac.edgecastcdn.net has address 68.232.35.121
cs91.wac.edgecastcdn.net has IPv6 address 2606:2800:234:124e:17ca:871:eb2:2067
koos@machiavelli:~$ telnet 0.gravatar.com 80
Trying 2606:2800:234:124e:17ca:871:eb2:2067...
Seems edgecast has an IPv6 reachability problem from here (home, xs4all IPv6) but not from other places. I can't reach the edgecast site either since it's on their own network. Or at least I can't via the squid proxy, I can when not using a proxy, since firefox has its own ideas about timeouts.

I tried to notify edgecast via e-mail and twitter.

Update 2012-11-12: Noted by others: WordPress IPv6 Issues - dereenigne.org seeing something more like PMTU problems.
And Edgecast is working on it: Thanks, @jmccrohan and @khoos. We're working hard on this right now. More soon. cc @florianoverkamp @cwoodfield
After a few changes and a lot of testing, we believe the IPv6 issue experienced by some users is now resolved.
But I'm still seeing routing issues:
$ traceroute6 www.edgecast.com
traceroute to www.edgecast.com (2606:2800:234:1df9:13d:1d4e:6b0:10cf) from 2001:980:14ca:61::13, port 33434, from port 44865, 30 hops max, 60 byte packets
 1  lo1.dr4.1d12.xs4all.net (2001:888:0:4401::1)  15.529 ms  14.870 ms  16.175 ms 
 2  1314.ae3.xr3.3d12.xs4all.net (2001:888:0:4403::2)  15.349 ms  16.757 ms  14.426 ms 
 3  0.ge-1-2-0.xr1.sara.xs4all.net (2001:888:1:4005::1)  17.594 ms  16.123 ms  15.212 ms 
 4  ge-0.ams-ix.amstnl02.nl.bb.gin.ntt.net (2001:7f8:1::a500:2914:1)  17.108 ms  16.405 ms  15.552 ms 
 5  ae-2.r03.amstnl02.nl.bb.gin.ntt.net (2001:728:0:2000::12a)  18.093 ms  17.441 ms  16.083 ms 
 6  * * *         
 7  * * *         
 8  * * *         
 9  * * *         

Update 2012-11-15: More traceroutes to test. It seems the only thing unreachable is exactly the /48 which holds the IPv6 content server(s).

Update 2012-12-07: A network engineer at xs4all contacted Edgecast and got the problem solved pronto. I guess he has a better entry point!
koos@greenblatt:~$ telnet -6 www.edgecast.com 80
Trying 2606:2800:234:1df9:13d:1d4e:6b0:10cf...
Connected to www.edgecast.com.
Escape character is '^]'.

Tags: , ,
2012-11-04 (#) 8 years ago
Some updates to the findings on playing with a cablemodem after getting access to a second cablemodem. The interesting conclusion is that the first one still seems to be 'blacklisted' as I can't get an answer to DHCP requests when using that one.

Tags: , , ,
2012-09-14 (#) 8 years ago
RIPE NCC Begins to Allocate IPv4 Address Space From the Last /8

IPv4 addresses aren't "finished" at RIPE yet, but allocation is now according to quite strict rules: IPv4 Address Allocation and Assignment Policies for the RIPE NCC Service Region.

Tags: , ,
2012-07-26 (#) 8 years ago
I run arpwatch on my home networks (wired and wireless are separate with slightly different policies). Since a Samsung Galaxy s3 showed up on the home wifi network things have been interesting. It does speak IPv6 with privacy enhancements. It also makes up a daily 10/8 address which get noted by arpwatch.

Tags: , , ,
2012-07-25 (#) 8 years ago
Internet history collected: the Internet history collection at the Computer History Museum in Mountain View, California. Close to El Camino Real, so near the places where the history happened.

Tags: , ,
2012-06-11 (#) 8 years ago
I recently got access to a Motorola SBV5121E cablemodem. No, I'm not switching to cable Internet access at the moment as ziggo still doesn't offer fixed IPv4 addresses or IPv6 at all on consumer connections. But it was interesting to give the cable modem a try.

If I am correct, this is the 'previous' generation cable modem. The maximum speed on the ethernet interface is 100 Megabit where Ziggo now offers subscriptions up to 120 Megabit.

When not connected to the cable network, the modem runs a small DHCP server for IPv4 which gives out 192.168.100.x addresses. The management interface is reachable via http://192.168.100.1/, username admin password motorola.

When I connect the RF interface to the tv cable, it gets a link. In the management interface I see Downstream frequency 186000000 Hz (186 MHz) with QAM256 modulation. The upstream frequency is 36000000 Hz (36 MHz) and the interface lists QPSK, 32QAM and 64QAM modulation, I guess the fastest one with a reasonable error-rate is chosen. Downstream speed is listed as 55.616 Megabit and upstream speed as 30.720 Megabit.

Interesting to see quadrature amplitude modulation (QAM) used here, just like in the DVB standards. I guess this is the best way to fit bits in a broadcast channel.

Connected to the ziggo cable and running tcpdump shows arp traffic originating from ziggo routers. Trying to cause traffic to ziggo IPv4 addresses from the outside shows arp for only one IPv4 address I tried, not for lots of others. Which means either all those clients are active (needing no ARP try needed) or one IPv4 address to the next being active in separate broadcast domains.

I also ran tcpdump with the modem starting up disconnected at the cable side. Absolutely no packet comes out on the ethernet interface.

First time I tried to get an IPv4 address from cable using DHCP while connected it gave me a IPv4 address from the 198.18.0.0/15 range, a range "reserved for benchmark tests of network interconnect devices". When I tried to visit any website I got redirected to www.ziggo-activatie.nl at 172.25.4.6 / 172.25.38.6 (which is in RFC1918 space). This website identifies itself as activation site for the provider, asking for a provisioning code and password. I guess the activation site will link cable mac and customer account in the databases at the cable operator side.

A scan of the assigned network (a /24 out of 198.18.0.0/15) shows several IPs in use all having the same MAC address 00:30:B8:xx:xx:xx (RiverDelta Networks). A search for RiverDelta Networks shows it is a maker of cable modem termination systems and has been acquired by motorola.

When I tried it again it with the same cable modem it doesn't give me an IPv4 address at all. An interesting change.

Update: Later another Motorola SBV5121E cablemodem found its way to me. Trying this modem gave me an IPv4 address via DHCP again and a new try at the activation server so I could look up certain things described in this item and update it.

Tags: , , ,
2012-06-06 (#) 8 years ago
IPv6 launch day going, with as first visible change for me, google advertises AAAA records even on networks not in the google ipv6 program:
koos@machiavelli:~$ host www.google.com
www.google.com is an alias for www.l.google.com.
www.l.google.com has address 173.194.67.103
www.l.google.com has address 173.194.67.104
www.l.google.com has address 173.194.67.105
www.l.google.com has address 173.194.67.106
www.l.google.com has address 173.194.67.147
www.l.google.com has address 173.194.67.99
www.l.google.com has IPv6 address 2a00:1450:4007:802::1012
Statistics to peek at: At this time (~ 11:30 CEST) there are no big peaks or rises in IPv6 traffic visible.

Update 2012-06-06 16:15 : A rise is visible in IPv6 traffic on the ams-ix.

Tags: , ,

IPv6 check

Running test...
, reachable as koos+website@idefix.net. PGP encrypted e-mail preferred. PGP key 5BA9 368B E6F3 34E4 local copy PGP key 5BA9 368B E6F3 34E4 via keyservers

RSS
Meningen zijn die van mezelf, wat ik schrijf is beschermd door auteursrecht. Sommige publicaties bevatten een expliciete vermelding dat ze ongevraagd gedeeld mogen worden.
My opinions are my own, what I write is protected by copyrights. Some publications contain an explicit license statement which allows sharing without asking permission.
Other webprojects: Camp Wireless, wireless Internet access at campsites, The Virtual Bookcase, book reviews
This page generated by $Id: newstag.cgi,v 1.34 2020/12/31 15:36:31 koos Exp $ in 0.022312 seconds.