Beta Test / Technical

[bdbcat]

Pavel....

1.3.6 is ancient, pre-git days. No quilting, no GL, no, no, no....

But, no Unicode, so it runs on Win95. I have it running on one of my everyday nav systems now, since I need W95 for some legacy stuff.

Amazing, almost 20 years old now....

Dave

[nohal]

Of course I know, I'm just surprised that it survived anywhere until now... I can't even count the machines that died for me since then.


Sent from

[fulup]

I made available an AIS Simulator application. OpenCPN is my reference platform for test. While it should work with any navigation chart system it works for sure at least with OpenCPN.

For quick test of AIS feed you only have to point OpenCPN onto my testing stream at tcp://sinagot.net:4001. For those who would like to customize route, have more boat, get special message flow, etc ... the application is available at https://www.npmjs.org/package/gpsdtracking

Note that the application provides a full AIS encoding library for the nasty 6bit binary payload. While I only implemented static and navigation report, implementing other message [Sart, Whether, Zone, etc ...] should be pretty trivial. The AIS Encode/Decode lib is written en nodejs, but as I mostly use OpenCPN source code as reference, who ever knows opencpn code should feel comfortable in encoding other messages. If needed I'm more than happy to provide help.

Note that for simpler test, I also provide basic GPRMC/NMEA & AIS/AIVDM a single vessel:
- tcp://sinagot.net:4002 NMEA/GPRMC
- tcp://sinagot.net:4003 AIS

[KenNorris49]

Some notes on using Opencpn 3.3.2118 with Ubuntu 14.04 32-bit

1. Excellent as it is, the bidirectional connection between my WiFi AIS unit and OCPN must be activated manually each time I open OCPN and again if (rarely) the connection is lost during use i.e. unchecked > Apply > checked > Apply

2. The floating windows of the toolbar and the dashboard occasionally become unresponsive i.e. the mouse cursor operates only on the underlying chart. The windows can be reactivated by clicking on the OCPN Launcher icon twice.

[muttnik]

Originally Posted by KenNorris49

1. Excellent as it is, the bidirectional connection between my WiFi AIS unit and OCPN must be activated manually each time I open OCPN and again if (rarely) the connection is lost during use i.e. unchecked > Apply > checked > Apply

Can you give a bit more info on the connection: I'm guessing this is TCP outbound from OCPN, receive and transmit checkboxes checked, but can you confirm? And are you pumping any info back towards the AIS unit or is it only actually receiving?

Normally I'd just go away and try and reproduce this but I'm completely flat out this week so if you can take a look at what's happening with tcpdump (i.e. tcpdump -i <interface> -l tcp port 10110 (or whatever)) that might be useful.

When OCPN drops a connection have a look with netstat to see if it thinks it still has the connection open, i.e. netstat -n --inet |grep 10110 (or whichever port you're using)

[KenNorris49]

Be delighted to track things down but please don't overestimate my competence.
The TCP connection is genuinely two-way:
both AIS (and GPS data if needed) directly from the Vespermarine XB8000 via the boat's internal WiFi setup and relayed data from the NMEA 0183 wind and depth devices; and
output from OCPN is of route data to a converter to the NMEA 2000(?) Simnet thence to the Simrad Autopilot (which eventually displays the data but doesn't acknowledge data to steer by, but that's another story).

So the Vespermarine XB8000 is at 192.168.1.2:39150

I'll try the tools you sent.

Ken

[KenNorris49]

So these are the outputs from netstat -n --inet |grep 39150

When active:

tcp 0 0 192.168.1.3:41930 192.168.1.2:39150 ESTABLISHED

When dropped without reconnection:

tcp 0 145 192.168.1.3:41930 192.168.1.2:39150 FIN_WAIT1

The connection is automatic at the outset iff the XB8000 is turned on when OCPN starts and I'm reasonably certain that is not invariably the case.

With sudo tcpdump -i wlan0 -l tcp port 39150

OCPN started before WiFi but with WiFi then activated:
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on wlan0, link-type EN10MB (Ethernet), capture size 65535 bytes

When the connection is reactivated:
ken@ken-HP-Pavilion-dv4-Notebook-PC:~$ sudo tcpdump -i wlan0 -l tcp port 39150
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on wlan0, link-type EN10MB (Ethernet), capture size 65535 bytes
08:23:53.065776 IP 192.168.1.2.39150 > ken-HP-Pavilion-dv4-Notebook-PC.41987: Flags [P.], seq 1793217152:1793217857, ack 2917995105, win 1448, options [nop,nop,TS val 9774 ecr 1730349], length 705
08:23:53.065803 IP ken-HP-Pavilion-dv4-Notebook-PC.41987 > 192.168.1.2.39150: Flags [.], ack 705, win 589, options [nop,nop,TS val 1730465 ecr 9774], length 0
08:23:53.066452 IP ken-HP-Pavilion-dv4-Notebook-PC.41987 > 192.168.1.2.39150: Flags [.], ack 705, win 589, options [nop,nop,TS val 1730465 ecr 9774], length 0
08:23:53.600599 IP ken-HP-Pavilion-dv4-Notebook-PC.41987 > 192.168.1.2.39150: Flags [P.], seq 1:73, ack 705, win 589, options [nop,nop,TS val 1730598 ecr 9774], length 72
etc.

[muttnik]

OK so OCPN knows the connection has dropped (or wasn't made) but it isn't reconnecting. I'm not going to get a chance to look at this for a few days but I will. Are you aware if previous versions reconnected as you expect (i.e., did I break something with the bi-directional thing or is this an enhancement opportunity)?

[KenNorris49]

I don't recall now. I think the two connections to WiFi, before you fixed it, were hit and miss and unreliable; but that might have more to do with my configuration than anything else.

[muttnik]

Having had a look it should reconnect and in my tests it is doing so without manual intervention. I'm using a slightly older version of the git source (from a few months back) and ubuntu 12.04. I'll update both when I'm on a better connection and have another look

[KenNorris49]

I just tried something else. Neglected to tell you that I have a second active connection to a GPS dongle via gpsd i.e. on 127.0.0.1:2947. That appears to block the WiFi connection; at least when I unchecked the gpsd connection and hit Apply the (your) WiFi connection happened immediately.
I have Priority 3 set for the WiFi and Priority 2 for gpsd.
Any tests?

[KenNorris49]

Tested your bi-directional WiFi connection every which way without the GPSD connection enabled and it seems to be bullet-proof — it reconnects without difficulty. But GPSD blocks the reconnection, irrespective of what order of priority for the connection, unless both sources are active when OCPN starts; in that circumstance, both connections operate as normal.

[benet250]

Hello, I have a problem that I do not know where to go with.
Hardware: Globalsat BT821C GPS receiver/Bluetooth transmitter, Asus TF810C (VivoTab) tablet upgraded from Windows 8.0 to 8.1 (32 bit only)with all updates installed. Dell Inspiron 1500 Laptop using windows 8.1 with all upgrades installed.
Software: OpenCPN, SeaclearII, Bluetooth 4.0
Symptoms: On the Dell (32/64 bit) when configuring OpenCPN comm on Data Port 4 at 38400 (or any speed) and then connecting (after pairing) to the BT 821C, the windows settings dialogue box changes to show Com3 incoming and Com4 outgoing with the Bluetooth device listed on both ports, and the dialogue box in OpenCPN shows two active com ports (Com3 and Com4 with output from Com4). Open CPN runs great as does SeaclearII
On the Asus TF 821C (32 bit only), when configuring OpenCPN same as the Dell, the windows settings dialog box shows Com3 as active but does not have a device listed, and Com4 is also active with the device listed, OpenCPN DOES NOT connect with the GS BT821C. SeaClearII works great however,
I have screenshots showing both Dell and Asus windows settings and OpenCPN settings.
Any ideas?.

[bdbcat]

benet250...

Please attach screenshots.

What version of OpenCPN?

Thanks
Dave

[benet250]

Hello
OpenCPN version is 3.2.2
here are attached screenshots