[arm-allstar] DNS Query

David McGough kb4fxc at inttek.net
Mon Aug 16 22:30:55 EDT 2021


Hi Doug,

Yes, it's fine to register with HamVoIP and ASL.  This is the best config 
to use, since it offers the best redundancy, should one or the other of us 
"have a bad day."  The order in which the servers are listed in the 
iax.conf file doesn't matter.

dns-query is a HamVoIP only command. In its most basic use, it queries
the HamVoIP DNS servers, scattered around to world, to retrieve the most
up to date connectivity info for the node of interest.  This data gets
cached briefly (typically 2 minutes or less), locally, on each system.  
Once the cache expires, an Internet lookup query is performed again.

The database which HamVoIP's DNS servers use is updated every 6 seconds
with data from both HamVoIP's registration servers and the ASL
registration nodelist.  Currently, this compilation of data is
continuously performed and distributed from two (or three) US data
centers, for geographic diversity.  Outages are very rare.

I suspect the issue you were having with the public IP address rapidly
changing was probably an ISP problem (likely CGNAT related?).  I've been
casually watching your node addresses over the last hour and haven't
noticed this issue, so far ??

As for rebooting your node, that typically isn't needed very often, when
using HamVoIP.  There are exceptions, however.  Sometimes various
applications (like apache or sshd) can fill up the /var/log filesystem,
leaving the node unstable.  On HamVoIP, /var/log is a tmpfs (RAM Drive).  
In this case, the quickest "fix" is simply to reboot, which empties
/var/log and resolves this issue.  There are also some long-standing
memory leaks in the currently released Asterisk/AllStar code.  Over time, 
this can cause out-of-memory issues.  A simple Asterisk restart (not a
full system reboot) will resolve this concern.  The latest pre-release
code has most of the app_Rpt memory issues resolved.  As an example of a
node that "has run for a while," I present this RPi3B.  This is a busy, 
wide coverage repeater.


[root at repeater-146820 ~]# uptime
 22:21:05 up 564 days,  5:25,  1 user,  load average: 0.02, 0.01, 0.00

[root at repeater-146820 ~]# asterisk -rx "core show uptime"
System uptime: 1 year, 17 weeks, 3 days, 1 hour, 31 minutes, 20 seconds 



73, David KB4FXC



On Mon, 16 Aug 2021, "Doug Theriault via ARM-allstar" wrote:

> I use dns-query command to monitor the IP address of several nodes that 
I manage via reverse SSH tunnels.  Custom script emails me if there's 
been a change in the IP address for the nodes router and a reverse ssh 
tunnel has dropped or is restored.

All my nodes/hubs are deployed using hamvoip distro, rPI3B+ based and 
use two registration lines in iax.conf; one to allstarlink.org, the 
other to hamvoip.org.  Allstarlink is listed first, followed by a line 
to register with hamvoip.

The script detects when an address has changed and automatically 
restores reverse ssh tunnel;  DHCP lease expiration is typically the 
cause, changing the IP address of the router for the node/hub.  To date 
its been working fine for quite some time.

As of Aug 11th, 16:55 I started seeing IP addresses on several nodes 
changing IP's. 4 to 5 / hr, peaked at about 20/hr couple days ago.  It 
seems to be slowing down but still occurs.  Flips between two IP 
addresses.  At first I suspected Vz, but given my config, I am wondering 
if its because there is a slight difference between the two registration 
services I am using.

I changed two of my nodes (ones which used Vz Cell routers) to use a 
simple "curl -s icanhazip.com" instead of dns-query and both those nodes 
IP addresses are stable using the curl method.  So I suspect I am 
getting different IP address when querying using dns-query.  Sometimes I 
get the NULL return which is fine, the script retries.  But the flip 
flopping IP change is baffling.

Questions:

1.  Am I creating a problem by registering nodes with both 
allstarlink.org and hamvoip ?

2.  Is there a write up somewhere on how dns-query works?  Does it 
utilize a file cache or does it query both allstarlink.org and hamvoip 
servers returning IP address of whichever server responds first?  Or is 
this command specific to hamvoip distro only?

3.  I have heard operators say they regularly reboot their nodes to 
prevent "problems".  I do not do this, I've had uptime on couple nodes 
of almost a year.  Is it recommended to regularly reboot a node to 
prevent any issues?


thanks in advance,

-- 
73,
Doug - NO1D

_______________________________________________

ARM-allstar mailing list
ARM-allstar at hamvoip.org
http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar

Visit the BBB and RPi2/3/4 web page - http://hamvoip.org



More information about the ARM-allstar mailing list