For those who want to join the map test with mikrotik routers:
- create a new script containing a single line (replace the call and locator, of course):
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=yo2loj@kn05os" keep-result=no
- in the scheduler get it to run every 5 minutes.
That's it.
ampr-ripd clients appear as green dots, MT routers as reddish ones.
Marius, YO2LOJ
Hi Marius...fails here I think because AMPRnet is not a default route/interface?
[admin@core100.wa4zlw.homedns.org] > /tool fetch mode=http url="http://44.182.21.1:59001/mikrotikid=wa4zlw@FN20bk" keep-result=no status: failed
failure: closing connection: <404 Not Found> 44.182.21.1:59001 (5) [admin@core100.wa4zlw.homedns.org] >
leon
On 6/1/2017 7:56 AM, Marius Petrescu wrote:
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=yo2loj@kn05os" keep-result=no
--- This email has been checked for viruses by AVG. http://www.avg.com
No, it fails because you forgot the '?' between 'mikrotik' and 'id'
On 01.06.2017 15:09, Leon Zetekoff wrote:
Hi Marius...fails here I think because AMPRnet is not a default route/interface?
[admin@core100.wa4zlw.homedns.org] > /tool fetch mode=http url="http://44.182.21.1:59001/mikrotikid=wa4zlw@FN20bk" keep-result=no status: failed
failure: closing connection: <404 Not Found> 44.182.21.1:59001 (5) [admin@core100.wa4zlw.homedns.org] >
leon
On 6/1/2017 7:56 AM, Marius Petrescu wrote:
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=yo2loj@kn05os" keep-result=no
AVG logo http://www.avg.com/internet-security
This email has been checked for viruses by AVG antivirus software. www.avg.com http://www.avg.com/internet-security
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
it's there the cut didnt get it
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=wa4zlw@FN20bk" keep-result=no
On 6/1/2017 8:10 AM, Marius Petrescu wrote:
No, it fails because you forgot the '?' between 'mikrotik' and 'id'
On 01.06.2017 15:09, Leon Zetekoff wrote:
Hi Marius...fails here I think because AMPRnet is not a default route/interface?
[admin@core100.wa4zlw.homedns.org] > /tool fetch mode=http url="http://44.182.21.1:59001/mikrotikid=wa4zlw@FN20bk" keep-result=no status: failed
failure: closing connection: <404 Not Found> 44.182.21.1:59001 (5) [admin@core100.wa4zlw.homedns.org] >
leon
On 6/1/2017 7:56 AM, Marius Petrescu wrote:
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=yo2loj@kn05os" keep-result=no
AVG logo http://www.avg.com/internet-security
This email has been checked for viruses by AVG antivirus software. www.avg.com http://www.avg.com/internet-security
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
--- This email has been checked for viruses by AVG. http://www.avg.com
That's so cool. I'd like to use such a map for other things, is this particular one standalone-able? Cheers
On 1 June 2017 at 12:56, Marius Petrescu marius@yo2loj.ro wrote:
(Please trim inclusions from previous messages) _______________________________________________ For those who want to join the map test with mikrotik routers:
- create a new script containing a single line (replace the call and
locator, of course):
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=yo2loj@kn05os" keep-result=no
- in the scheduler get it to run every 5 minutes.
That's it.
ampr-ripd clients appear as green dots, MT routers as reddish ones.
Marius, YO2LOJ
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
The map itself is standalone (you just need to get the html file and the java scripts that are loaded at startup - see in 'view source').
The JSON event source is a dedicated daemon which I have written for this purpose.
Marius, YO2LOJ
On 01.06.2017 15:17, Kathie Dart wrote:
(Please trim inclusions from previous messages) _______________________________________________ That's so cool. I'd like to use such a map for other things, is this particular one standalone-able? Cheers
Guys, when adding the script to the routers, please edit the callsign and locator BEFORE running it.
On 01.06.2017 14:56, Marius Petrescu wrote:
(Please trim inclusions from previous messages) _______________________________________________ For those who want to join the map test with mikrotik routers:
- create a new script containing a single line (replace the call and
locator, of course):
/tool fetch mode=http url="http://44.182.21.1:59001/mikrotik?id=yo2loj@kn05os" keep-result=no
- in the scheduler get it to run every 5 minutes.
That's it.
ampr-ripd clients appear as green dots, MT routers as reddish ones.
Marius, YO2LOJ
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
Ok, since it seems taht there is already some spoofing going on, I will restrict the senders for mikrotik routers to 44net sources only.
This should have no effect on legitimate users.
Marius, YO2LOJ
On 01.06.2017 16:08, Marius Petrescu wrote:
(Please trim inclusions from previous messages) _______________________________________________ Guys, when adding the script to the routers, please edit the callsign and locator BEFORE running it.
I'm more inclined to think it's misconfigured routers rather than deliberate spoofing, but you may be right. - Brian
On Thu, Jun 01, 2017 at 04:16:04PM +0300, Marius Petrescu wrote:
Ok, since it seems taht there is already some spoofing going on, I will restrict the senders for mikrotik routers to 44net sources only.
Unless there are valid nodes with non-existing callsigns on invalid QTH locators registernig from the internet, then maybe.
On 01.06.2017 17:29, Brian Kantor wrote:
(Please trim inclusions from previous messages) _______________________________________________ I'm more inclined to think it's misconfigured routers rather than deliberate spoofing, but you may be right.
- Brian
On Thu, Jun 01, 2017 at 04:16:04PM +0300, Marius Petrescu wrote:
Ok, since it seems taht there is already some spoofing going on, I will restrict the senders for mikrotik routers to 44net sources only.
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
But how come that this list leak to non ham persons that the know what to do ?
________________________________
You must be right. Too bad, I like to think better of people. - Brian
I assume it's because some number of the subscribers to the mailing list are bots or untrustworthy people. Or maybe they're archive services or remailers that make the list available to the general public.
Consider the fact that every time I post a URL that points at a server whose logs I can watch, there is an immediate attempt to fetch that URL. It happens too quickly to be real people, it's got to be a bot of some sort. And there isn't just one of them, there are typically three or four from different IP addresses in the first few minutes. - Brian
On Thu, Jun 01, 2017 at 06:31:08PM +0300, Marius Petrescu wrote:
Now this is really a good question. On 01.06.2017 18:16, R P wrote:
But how come that this list leak to non ham persons that the know what to do ?
Hey Brian,
I imagine we have a few possible solutions:
- high impact: delete all users from the email list and make everyone re-subscribe. I don't know how effect this will be unless the Postman system can enable something like a captcha challenge system to keep out bots. People who don't keep up with their email would loose some of the postings
- no impact : When anyone posts a URL, etc. break them which will require some human manipulation. For example
HOST=44.182.21.1 http://%5Bhost-from-above%5D:59001/mikrotikid=n0call@AA20aa
This won't be perfect as not everyone will remember to do this, they will post URLs that they use but might impact other user's infrastructure (like Marius's new tool)
- Maybe there are some other, better ideas..
--David KI6ZHD
On 06/01/2017 08:46 AM, Brian Kantor wrote:
(Please trim inclusions from previous messages) _______________________________________________ I assume it's because some number of the subscribers to the mailing list are bots or untrustworthy people. Or maybe they're archive services or remailers that make the list available to the general public.
Consider the fact that every time I post a URL that points at a server whose logs I can watch, there is an immediate attempt to fetch that URL. It happens too quickly to be real people, it's got to be a bot of some sort. And there isn't just one of them, there are typically three or four from different IP addresses in the first few minutes.
- Brian
On Thu, Jun 01, 2017 at 06:31:08PM +0300, Marius Petrescu wrote:
Now this is really a good question. On 01.06.2017 18:16, R P wrote:
But how come that this list leak to non ham persons that the know what to do ?
Mailman doesn't have CAPTCHA. It does use a challenge; when you request to subscribe, it mails you a key that you have to mail back, but I know that bot subscribers have already automated that. It helps keep some undeliverable addresses from subscribing, but it's not designed as a real security measure.
I've considered using a different mailing list manager, but I haven't found one that was superior enough to make it worth the hassle to transfer the large list of existing subscribers. And I guarantee that if people had to resubscribe, a lot of them wouldn't.
No, I just assume that anything posted to the list becomes public knowledge. The only solution is not to post secrets. - Brian
On Thu, Jun 01, 2017 at 10:55:04AM -0700, David Ranch wrote:
- high impact: delete all users from the email list and make everyone
re-subscribe. I don't know how effect this will be unless the Postman system can enable something like a captcha challenge system to keep out bots. People who don't keep up with their email would loose some of the postings
Brian Kantor said:
Mailman doesn't have CAPTCHA.
Here's a patch for adding reCaptcha support to Mailman:
https://www.dragonsreach.it/2014/05/03/adding-recaptcha-support-to-mailman/
It does use a challenge; when you request to subscribe, it mails you a key that you have to mail back, but I know that bot subscribers have already automated that. It helps keep some undeliverable addresses from subscribing, but it's not designed as a real security measure.
My standard practice (on all Mailman instances I admin) to prevent "drive-by spamming" is configuring lists so that new subscribers are moderated and checking their initial submissions. It does require a bit of the moderator's time, but Mailman's WebUI makes the process quite easy.
Moderation is a good tool for weedinf out problem subscribers. Use the control at the bottom off /mailman/admin/<list-name>/members to turn on/off everyone's moderation bit.
73,
Thanks, Steve. That looks useful, but will take some effort to adapt it to FreeBSD which is what the mailing list host is running (not Linux). - Brian
On Thu, Jun 01, 2017 at 03:27:13PM -0500, Steve K0STK wrote:
Here's a patch for adding reCaptcha support to Mailman: https://www.dragonsreach.it/2014/05/03/adding-recaptcha-support-to-mailman/
Brian Kantor said:
Thanks, Steve. That looks useful, but will take some effort to adapt it to FreeBSD which is what the mailing list host is running (not Linux).
You'll need to install www/py-recaptcha
http://www.freshports.org/www/py-recaptcha/
And then update lines 17 and 40 in the patch to reflect the location of pyshared on FreeBSD (probably /usr/local/share/pyshared)
That should be it.
The FreeBSD ports package 'py-recaptcha' says that you need to obtain a key from http://recaptcha.net/api/getkey. Unfortunately, that URL and shorter variations on it just return 404. That domain is served by Google but appears to be empty. There's certainly nothing there anymore. Apparently you now have to sign up with Google and get a private key from the Google developers program.
Why is nothing simple anymore? - Brian
On Thu, Jun 01, 2017 at 03:27:13PM -0500, Steve K0STK wrote:
Here's a patch for adding reCaptcha support to Mailman: https://www.dragonsreach.it/2014/05/03/adding-recaptcha-support-to-mailman/
Marius,
A few questions about the dynamic map:
1) How long does it take for an entry to appear on the map? I manually entered the following URL into a browser (three times) but I don't see it on the map (CA, USA). http://44.182.21.1:59001/generic?id=n6mef@cm97aj
2) How long does an entry remain on the map after the last time you heard from it? Dynamic is nice, but if a gateway is down for maintenance, it would still be good to know that it exists.
3) Follow-up to #2: Most similar applications (winlink2K, irlp) use the colors or status tables to represent status (green=up, red=down, etc.). That would be more useful to me as a sysadmin than how the information was learned (which I realize is useful to you). For example, colors could represent age of information: green="heard from within the last hour", yellow="... within the last day", red="... within the last week". After that, if falls off the map. That would give us a hint as to whether the machine is more likely to be temporarily or permanently down. Would you consider?
4) Have you given any thought to Brian's suggestion that the input be in the DNS LOC record format, rather than callsign@maidenhead? It seems like a good way to use your infrastructure while preparing for integration with more "standard" tools later.
Michael N6MEF
Hi Michael,
On 02.06.2017 23:13, Michael Fox - N6MEF wrote:
Marius,
A few questions about the dynamic map:
- How long does it take for an entry to appear on the map? I manually
entered the following URL into a browser (three times) but I don't see it on the map (CA, USA). http://44.182.21.1:59001/generic?id=n6mef@cm97aj
The entry appear immediately on the map (on reception, the daemon sends out a JSON event stream element to all listeners ASAP - less the 1 sec) .
- How long does an entry remain on the map after the last time you heard
from it? Dynamic is nice, but if a gateway is down for maintenance, it would still be good to know that it exists.
For the moment it stays on the map indefinitely (I still have to figure out how to delete a marker). In the internal buffer of the daemon it will be there for 1 hour and will be available on refresh for that time. But this can be changed anytime since it is just a define in the code.
- Follow-up to #2: Most similar applications (winlink2K, irlp) use the
colors or status tables to represent status (green=up, red=down, etc.). That would be more useful to me as a sysadmin than how the information was learned (which I realize is useful to you). For example, colors could represent age of information: green="heard from within the last hour", yellow="... within the last day", red="... within the last week". After that, if falls off the map. That would give us a hint as to whether the machine is more likely to be temporarily or permanently down. Would you consider?
This is just a starting point. We can do whatever we want. But again, to change the colors, I need to find out how to modify existing map markers. Still ToDo.
- Have you given any thought to Brian's suggestion that the input be in
the DNS LOC record format, rather than callsign@maidenhead? It seems like a good way to use your infrastructure while preparing for integration with more "standard" tools later.
As I said, it is our project, we can do whatever we need. I just wanted a quick start, to keep the momentum. Next steps are wide open from my side.
Addendum:
1) How long does it take for an entry to appear on the map? I manually
entered the following URL into a browser (three times) but I don't see it on the map (CA, USA). http://44.182.21.1:59001/generic?id=n6mef@cm97aj
For entries to be accepted, the sender needs to be in the 44net address space. Otherwise http notifications will be ignored. This could be a possible cause.