Same client can connect one minute, but is "UNAUTHENTICATED" on second attempt

Folks - I’m a new user, working to get node 60268 working correctly. I have been struggling with two possibly unrelated issues:

(1) When the node is authenticated (both shown locally, and as per the authenticated=YES flag on my local node information on site), I am often able to connect via my iPhone (repeaterphone) client or my Mac (Transceive) client the first time… but not on a second attempt just moments later.

In looking at the CLI debug spew, it’s clear that the second attempt comes in as “UNAUTHENTICATED” and is rejected.

Some clues seem to be that if I reset the auth info in the client, that I can sometimes reconnect. This makes me feel that, somehow, only one client can be authenticated at a time - anyone know about this? Any other theories for why this might happen?

(2) On a reboot, the server seems not to correctly register with That is, the local server claims it’s registered (CLI “iax2 show registry”) but the website claims the node is not registered. Some time later, maybe without my doing anything, the issue is resolved.

In another thread, I read that somehow the local server can claim registered because there is no reliance on the incoming port forwarding on 4569 to work for this process. It’s unclear whether the incoming link would make the allstarlink site report not registered, or if it might show registered but incoming connections would fail. Help on this one?

I am seeing the dns entry for seemingly flop between a couple addresses, and after a dns refresh, again, my server seems to be able to re-register. Here’s the event:

— snip —
K6FB-allstarCLI> dnsmgr status
DNS Manager: enabled
Refresh Interval: 300 seconds
Number of entries: 1
CLI> dnsmgr reload
[Apr 28 10:37:56] NOTICE[3948]: dnsmgr.c:394 do_reload: Managed DNS entries will be refreshed every 300 seconds.
== Refreshing DNS lookups.
[Apr 28 10:40:03] NOTICE[3889]: dnsmgr.c:178 dnsmgr_refresh: host ‘’ changed from to
== Manager ‘admin’ logged on from
— /snip —

*** Update: website shows not registered, but CLI iax show registry shows registered, and an inbound public auth call shows as AUTHENTICATED but still doesn’t accept the connection. Here’s the debug:

— snip —
K6FB-allstar*CLI> iax2 show registry
Host dnsmgr Username Perceived Refresh State Y 60268 179 Registered
– Accepting AUTHENTICATED call from
> requested format = ulaw,
> requested prefs = (ulaw),
> actual format = ulaw,
> host prefs = (ulaw|gsm),
> priority = mine
– Executing [s@allstar-public:1] Ringing(“IAX2/allstar-public-2011”, “”) in new stack
– Executing [s@allstar-public:2] Set(“IAX2/allstar-public-2011”, “RESP=???”) in new stack
– Executing [s@allstar-public:3] Set(“IAX2/allstar-public-2011”, “NODENUM=60268”) in new stack
– Executing [s@allstar-public:4] GotoIf(“IAX2/allstar-public-2011”, “1?hangit”) in new stack
– Goto (allstar-public,s,15)
– Executing [s@allstar-public:15] Answer(“IAX2/allstar-public-2011”, “”) in new stack
– Executing [s@allstar-public:16] Wait(“IAX2/allstar-public-2011”, “1”) in new stack
– Executing [s@allstar-public:17] Hangup(“IAX2/allstar-public-2011”, “”) in new stack
== Spawn extension (allstar-public, s, 17) exited non-zero on ‘IAX2/allstar-public-2011’
– Hungup ‘IAX2/allstar-public-2011’
— /snip ----

There is a clue here in that as I was monitoring the CLI debug spew, I watched something reset in the dynamic DNS and asterisk reported some change of DNS entry for something… and immediately after this, I was able to connect and the site reported registered. This leads me to think that my issue is wrapped up in dnslookups – but I didn’t think incoming connections relied on this, as I thought it was allstarlink’s system that knew my IP/port and directed traffic there. I read that it is expected behavior for the dns servers to rotate between five different servers and that any one of them should work (i.e. a stale DNS should be fine, and only a problem for 15min until refreshed).

Clearly, I’m a new user and still unclear on how the system works. Hoping a patient soul will offer me some guidance here. I think I’m close. I have allmon2 issues, too, but I’ll get help on that after the server at least runs reliably.

Many thanks!

Brian AG6WR for node 60268, K6FB linked repeater system, silicon valley, CA

I’m not sure exactly; my thoughts are thus: There are delays in the system (nodes, servers, web pages). I’ve noticed if changes are made to quickly one can get into modes where it seems things are not working correctly because the system state as reported by the web page is behind. In flying, we call the phenomenon PIO (pilot induced oscillations).