Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - g4swx

Pages: 1 2 3 [4]
46
General discussion forum / Re: Stumped on Audio Drops
« on: 2011-10-21, 12:31:29 »
It is also highly likely that the drop outs could be caused by bursty Internet traffic to/from either end of your link.

Typical causes of this are webcams, particulary at the radio end, Outlook like mail clients at either end and IPTV applications at the control end. You must limit the upstream bandwidth of all clients at the radio end, except possibly the remote rig itself. In particular I would recommend setting an upsream limit of 100Kbps (unless you have >1Mbps upstream bandwidth) on all other clients on that network. On the control end if you have ~10Mbps then set the limits at around 1Mbps. Note this will 'slow' access to the Internet for other usage and may well stop most IPTV applications from working.

Sometimes you will find particular routers, particularly the cheaper ones(even Apple!) are part of the problem. If all else fails try changing the upstream and then the downstream router.

Rx jitter/delay settings of 10 are often not enough, I have to run with 16/15 AND I had to reduce the audio quality setting to 1 to maintain a basic link between two different ISPs.

Even with an all private network it is not easy to elimate all such drop out except by the introduction of rigourous QoS limiting the bandwidth that can be eaten up by other systems on the network. HTTP traffic can be a real pain and should be constrained!

Sometimes I can also cause such a drop-out by sending an HTTP request to the web server on remote rig (radio) box!

Fixing drop outs just comes down to paying very close attention to what is running on your IP networking(assuming the ISP connection is OK). I would also recommend using pingtest.net for testing your ISP/router.

73
John G4SWX

47
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-07-22, 13:29:17 »
Sorry for the delay - I got my ADSL drop-outs fixed and this became less of an issue.
However I have now captured all of the packets in the failure mode and although I had thought it was a DNS cache issue in the RRC1258 it is the the router holding onto the NAT session for the DNS look-up! Much as you suggested this is obviously a router bug but with this particular router Linksys WAG300N it only seems to impact DNS. It also times out in about 5 mins.....
So; the DSL drops, router gains new external IP address whilst still 'hot', attempt to connect RRC-1258, DNS look-up fails........ Checked this by configuring a PC in a similar manner.
Complete fix = trash the Linksys router. :-[
Thank you for your help :)
73
John G4SWX

48
Hardware, Cabling, Installations / Re: OUT0 mode "connect"
« on: 2011-07-22, 13:11:33 »
I too have had this problem - I have been using OUT0 to drive a DL4KH sequencer where the 'gnd on TX' input pulls a PNP transistor circuit down. The input is held up to 12V with a 2.2K resistor so that the 8-9V via D10 easily switched it over. As a 'quick fix' I added a series resistor to limit the pull-down but now that I understand the issue I will probably use a 4.7V Zener diode in series with the sequencer input.
73
John G4SWX

49
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-06-02, 21:44:32 »
> This thing you describe is still not possible to understand. DHCP or not DHCP, when the RRC is behind > a NAT router it should not cease functioning after in your case a disrupt in the external network
> (ADSL).

I would tend to agree on the theory but it will be dependant upon what the IP stack puts into cache and how it manages its (internal network) DHCP handshake. It seems as though every other device on my network, PCs(Win + LINUX), IPhones, Internet Radio, 2 x STB etc, fixed or DHCP has no problems with the router (Cisco- Linksys) of the loss of ADSL + change of supplier IP. Even changing the router from Linksys to 2Wire does not resolve it.

Using DHCP for the RRC controller is not really acceptable as if there is any issue, and there are several every day) I first have to hunt in the router's DHCP table and guess what IP it has.

The problem is probably triggered more often as in the UK the ISPs circulate their IP addresses, from a large pool of quite different /24s on the DHCP servers. That means that if your router drops its PPP session for whatever reason it will almost always gain a new external address often from a very different netblock. This means that the router NAT table has to be refreshed with each new request from a downstream device. It seems that the RRC IP stack does not behave in the same way as all of the other devices on my network. It does, like all other devices retain the same (internal) IP address after such a change. Perhaps it is a quirk of the embedded stack and the caching of the DHCP credentials, in particular when an 'external' DNS server has been selected. However a 10 (not 5!) minute power-down will reset things.
If I have time I will capture 'normal' and 'lock-up' traffic with Wireshark.
73
John G4SWX

50
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-05-31, 21:27:04 »
OK, - I think that I have found a way around the problem:
I had set up an IP configuration on the control RRC that included; IP address, subnet mask, gateway address and external DNS server. Howevever if I delete all of these and set the IP setting to DHCP the problem goes away. ie when the ADSL at the control station drops momentarily, restores but with another IP address from the ISP DHCP pool the RRC continues, after the obvious drop-out, to work. However if I set an external DNS server address and the ADSL drops, the RRC ges into the "DNS Status Error.." condition and the only cure is to power down the RRC for ~ 10mins.
What is curious is that I have changed the ADSL router from Linksys to 2Wire and can repeat the problem and the work-around.
So the question comes back to one of what network information gets cached in the RRC as it should not be aware of the router NAT session?
73
John G4SWX

51
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-05-19, 23:06:57 »
It is still hapenning, at least once per day. I have set a fixed IP 192.168.1.125 for the control RRC so it is on the same subnet but free from DHCP issues at the router.
When the blip on the ISP side occurs I usually note a different  DHCP (ISP) address on the router WAN side. But as I am running a fixed IP for the RRC on the LAN side (Ethernet not wireless) this should not matter.
The DNS status then goes to error - failed to resolve
With power applied I have rebooted 4 or 5 times - the RRC first comes up DNS status : unknown
but on attempting a connection it fails with the DNS stutus in error - failed to resolve....
Powering down for a minute or so is not enought - when attempting a connection it fails with the DNS stutus in error - failed to resolve....
After about 8-10 mins of no power the RRC boots-up, initially with DNS status: unknown but after about 30 secs the status reports Ok!
I am running software 2.35 - would an update help? or is it possible that have I problem with MAC caching on the router(Linksys) rather than caching on the RRC?
However if it was a router issue I would expect to have an issue with my PCs which I do not, even when they are on DHCP.
73
John G4SWX

52
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-05-16, 16:09:37 »
> How long is a "break in the ISP connectivity"?
Can be as little as 20secs but mostly no more than 2 mins
>And when the "DNS Status Error.." occurs, does it remain as "Error" not only for the duration of >the "break", but also after the connection has been restored?
That is the point I am attempting to make, an interuption during the break is obvious but it remains after the break is restoration. Disruption of the DSL connection causes the RRC to get hung in a mode where a simple 'reset' will not sort it out, this can remain for up to several hours (at which power down everything). Removing the power from the RRC for ~ 10 mins and then restarting cures the problem.
>Also, do you mean that you don't have NAT at the router, so that the control RRC gets a new
>(external) IP address after each break(if on DHCP)?
Yes I do use NAT and DHCP at the router so that the RRC should, in theory, see nothing of the external network. Yet it seems to be aware of the connectivity failure.
That is why I asked about the DNS cache. Perhaps I should be asking about the cache of all of the network credentials as I guess that is what gets removed in a 10min power down.
73
John G4SWX

53
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-05-13, 12:35:41 »
I looked at the configuration between the RRC-1258 controler IP settings and the ADSL router(Linksys).
If I select DHCP (which I would prefer) as soon as I get a break in the ISP connectivity I almost instantly get the DNS error desribed below on the controler RRC-1258. No matter what DNS server has been selected.

If I select a fixed IP address and an external DNS server, Google etc, I still get the error:

DNS status Error, failed to resolve xxxxxxx.ham-radio-op.net!

Resetting the RRC-1258 several times does not cure the problem - however removing its power for 10 mins does. So my original thoughts concerning DNS caches remain

John G4SWX

54
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-05-11, 18:55:46 »
I don't think that is down to the ISP, I selected Google DNS on 8.8.8.8 and still had the same occasional problem. I will try setting the DNS server as the ADSL gateway
I also had an issue with the remote failing to update with DynDns (error 8 - I think) which remained until I remotely rebooted the RRC.
73
John

55
General discussion forum / Re: RRC-1258 DNS Cache
« on: 2011-05-10, 21:42:50 »
More - I have just suffered the same problem

The Radio side is fine (actual IPs changed):
Status
Name Value
Radio ON
SIP status Error
Last SIP error SIP Error
RTP Timeout 0
SIP Timeout 0
Rx Jitter buffer size 4
Rx Jitter delay 3
External IP xx.yy.164.232
SIP Out port 6060
SIP In port 6060
Audio Out port 1029
Audio In port 11500
Command Out port 1029
Command In port 12500
External SIP In port 6060
External Audio In port 11500
External Cmd In port 12500
Other party xx.yy.50.217
Input 1 High
Input 2 High
Output 0 Low
Output 1 Low
Output 2 Low
DynDns status OK
Ping status (watchdog) Off
Active profile: Default
PTT status: OFF

However the controller has lost the DNS status because of a network break at the controler end:

Status
Name Value
Control panel ON
Radio OFF
SIP status Error
Last SIP error SIP Error
RTP Timeout 0
SIP Timeout 0
Rx Jitter buffer size 16
Rx Jitter delay 3
SIP Out port 6060
SIP In port 6060
Audio Out port 11500
Audio In port 11500
Command Out port 12500
Command In port 12500
External SIP In port 6060
External Audio In port 11500
External Cmd In port 12500
Other party xx.yy.164.232
Input 0 High
Input 1 High
Input 2 High
Output 0 High
Output 1 Low
Output 2 Low
DNS status Error, failed to resolve xxxxx.ham-radio-op.net!
Active profile: Default
PTT status: OFF

Simply rebooting the controler is not enough.

John G4SWX


56
General discussion forum / RRC-1258 DNS Cache
« on: 2011-05-10, 18:52:29 »
My overall experience of the RRC is great.
However I have experienced a strange DNS failure a few times and I am wondering if it has something to do with the length of time that the RRC-1258 caches DNS. I am currently running 2.35 firmware and a dyndns address.

I have recently been suffering from intermittent operation on my control ADSL connection. As a result the IP address changes to another from the DHCP pool. You would think that this would not matter but after the customary connection failure when I try to reconnect from the control RRC to the radio and the connection attempt fails. The control RRC status indicates that the DNS status is unresolved... However with a PC on the same LAN I have no issues in connecting to the radio webserver even if I clear the PC's DNS cache and start again. The only difference as far as I can see is that I have entered my ISP's primary DNS server in the control RRC IP settings whereas the PC gets the DNS server address from by DHCP from my local router.
It does resolve, but not very quickly

Any thoughts?
John G4SWX

Pages: 1 2 3 [4]