Article: 51192 of rec.radio.amateur.digital.misc From: devnull@ec.rr.com (Mark White) Subject: Re: do I have to rebuilt the kernel to enable AX25? References: <1138479213.573357.165960@z14g2000cwz.googlegroups.com> Message-ID: Date: Sat, 28 Jan 2006 20:53:47 GMT http://www.tldp.org/HOWTO/AX25-HOWTO/index.html Article: 51193 of rec.radio.amateur.digital.misc From: Bob Bob Subject: Re: Can't run TCP/IP through AX.25 using Linux References: <1138719759.587807.123920@g43g2000cwa.googlegroups.com> <1138825794.019667.188990@g47g2000cwa.googlegroups.com> Message-ID: Date: Thu, 02 Feb 2006 10:38:53 -0600 Okay I see.. problems problems! When I first started playing with TCP/IP over packet maybe 12 years ago we did all we could to stop ARP broadcasts in our local subnet. All of the users had static ARP tables for that reason. We also had static host lookups. More recently I was involved in an IP over satellite paging/short message system where our initial bandwidth allocation was very small (3kb/sec). I made a lot of noise over that setup and suggested that if the b/w was unchangable then we should use static ARP tables. They increased the b/w instead, as a manual maintenance of an IP networks ARP's is a huge task! If you arent intent on using busybox you may want to try the Knoppix run >from CD version that has the amateur radio add-ons. re PPP and HDX/FDX. Well in a way you are simply shifting the HDX switching back to the computer instead of using the radio hardware. I'd suspect that in the end you wouldnt do any better. I am however prepared to be shot down over this! I wonder if SLIP is any better handling HDX? It has less protocol overhead as well.. (from memory) Not sure about compression though. I have an application of sorts for this same mind of thing. My mother in law lives about 500 yards away just over the brow of a hill. There are also lots of trees! I doubt 2.4GHz will work and was looking at something with at least 115K throughout to connect her PC to our network. What has been your experience with the 900MHz gear and how well does it do with path obstacles? Oh and what did the units cost? Tnxs for the info on tcpdump. One day when I get back to radio... Cheers Bob W5/VK2YQA East Texas gfilion@gmail.com wrote: > > Bob, thanks a lot for your help. Article: 51194 of rec.radio.amateur.digital.misc From: Joseph Fenn Subject: Re: Can't run TCP/IP through AX.25 using Linux Date: Sat, 4 Feb 2006 15:41:55 -1000 Message-ID: References: <1138719759.587807.123920@g43g2000cwa.googlegroups.com> This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. --1465796871-877555047-1139103715=:21804 Content-Type: TEXT/PLAIN; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Dont know about you guys, but since I still use an old 8 bit Commodore128 computer I do all my packeting with the software DIGICOM128 AND IT WAS WRITTEN BY THE GERMANS WAY BACK. I use only the VHF band for military packet relays from Ft Huachuca Az to Schofield Bks here in Honoululu. Joe ABM6JF ********************************************************** * Ham KH6JF AARS/MARS ABM6JF QCWA WW2 VET WD RADIO SYSTEM* * Army MARS PRECEDED by AARS (Army Amateur Radio System) *=20 * Hi State ARMY MARS COORDINATOR * ********************************************************** On Sat, 4 Feb 2006 gfilion@gmail.com wrote: > Hi again, > > I did some tcpdump and it definitly looks like an ARP problem. > > root@ali:~/tcpdump-ax25/tcpdump-3.9.3# ./tcpdump -envi ax0 > tcpdump: listening on ax0, link-type AX25 (AX.25), capture size 96 > bytes > > [... root@slug:~# ping 172.25.25.1 ...] > 17:30:01.703620 VA2JF-2 > -6: Res 3, C/R 0, UI, ?, pf 0, length 101: > (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto: ICMP (1), length: > 84) 172.25.25.2 > 172.25.25.1: ICMP echo request, id 1691, seq 0, > length 64 > [... Notice the "VA2JF-2 > -6" instead of "VA2JF-2 > VA2JF-1" ...] > > [... root@ali:~# ping 172.25.25.2 ...] > 17:31:24.033354 VA2JF-1 > VA2JF-2: Res 3, C/R 0, UI, ?, pf 0, length > 101: (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto: ICMP (1), > length: 84) 172.25.25.1 > 172.25.25.2: ICMP echo request, id 6003, seq > 1, length 64 > [... I never get an answer from slug, I guess it's because slug's arp > table doesn't contain ali's address ...] > > I'll get in contact with the creator of this kernel. > > 73 > GFK's > -- > Guillaume Filion, ing. jr > Logidac Tech., Beaumont, Qu=E9bec, Canada - http://logidac.com/ > PGP Key and more: http://guillaume.filion.org/ > > --1465796871-877555047-1139103715=:21804-- Article: 51195 of rec.radio.amateur.digital.misc From: Bob Bob Subject: Re: Can't run TCP/IP through AX.25 using Linux References: <1138719759.587807.123920@g43g2000cwa.googlegroups.com> <1138825794.019667.188990@g47g2000cwa.googlegroups.com> <1139087448.093090.220320@g14g2000cwa.googlegroups.com> <1139094572.120236.18370@g14g2000cwa.googlegroups.com> Message-ID: Date: Sun, 05 Feb 2006 15:54:50 -0600 I wonder if slug has an issue with dashes in the hardware address? Either that or it is smart enough to notice the first 5 chars of the hardware address are the same source > dest. As you say though it isnt working! Looks like you are getting closer! Okay on the 50:50 half duplex timing of the boxes. That seems kind of a strange way to do it. Would have expected the manufacturer to use dynamic b/w allocation. I had a quick look at SLIP/CSLIP but didnt really establish how effective it was on HDX. I assume that ppp is unusable because of the backgound handshaking that goes on (ipcp etc)? Is it that or timeouts and retries at the IP layer? Tnxs for the info on your purchase. I dont want to spend lots of dollars on my attempt. Would rather invest in 2nd hand than new equipment. The MIL is using dial-up at the moment so it isnt exactly urgent. I have a mind to use one of the scrap microwave radios from work but they start at 2.4GHz. (We manufacture data equip up to 12GHz and DS1 through E3) My path actually looks a lot better than yours though! Cheers Bob gfilion@gmail.com wrote: > > Hi again, > > I did some tcpdump and it definitly looks like an ARP problem.