NMEA output to TCP/UDP is slow and incompatible

[cwensley]

First I just want to say thanks for all the hard work on this! I use vyacht regularly and it works great.

I recently upgraded from 1.3.27 to 1.3.30 and noticed that the stream from UDP or TCP only emits every 4-5 seconds. On some instrument readouts (e.g. OnSiteWX ), it'll show it as NO DATA quite often because of this. Furthermore, the wind speed reads as 999.0, whereas with 1.3.27 it read correctly.

Do you know what's going on there?

Also, (as an open source developer I hate these but there is no other 'forum' for vyacht that I know of) are there any instructions for building a new custom firmware? I would like to contribute to this project to fix some of the bugs that I'm running into, update instrument panel (v2 works WAY better), etc. A single repo that pulls in the right stuff via submodules and includes the build scripts would be excellent. If you don't have time for that, I don't mind helping out if I can.

Thanks, and keep up the good work!

[vyacht]

Hi Curtis sorry for not noticing this earlier. Spam filter too aggressive.

On 23 Feb 2018, at 21:15, Curtis Wensley ***@***.***> wrote: First I just want to say thanks for all the hard work on this! I use vyacht regularly and it works great.

Great to hear.

I recently upgraded from 1.3.27 to 1.3.30 and noticed that the stream from UDP or TCP only emits every 4-5 seconds. On some instrument readouts (e.g. OnSiteWX ), it'll show it as NO DATA quite often because of this. Furthermore, the wind speed reads as 999.0, whereas with 1.3.27 it read correctly. Do you know what's going on there?

It would be very helpful if you had a log from the router. Either you can go to the router’s shell and start the gpsd process with the -N -D 5 options in addition to the other options. Or you just use http://<routerip>/test.html

Also, (as an open source developer I hate these but there is no other 'forum' for vyacht that I know of) are there any instructions for building a new custom firmware? I would like to contribute to this project to fix some of the bugs that I'm running into, update instrument panel (v2 works WAY better), etc. A single repo that pulls in the right stuff via submodules and includes the build scripts would be excellent. If you don't have time for that, I don't mind helping out if I can.

Thanks for your offer! Its really a time-issue for me. The gpsd repo (which contains a heavily modified version of gpsd for the router’s purpose) is more or less up-to-date. Other parts of the router software are not. I will upload them as tgz to start with. Unless you have a better idea of cause. Best Regards Bernd

…

Thanks, and keep up the good work! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[cwensley]

Hey Bernd,

Thanks for the response! Here's the log output for about 30 seconds each. It appears that the data is missing for quite a few of the entries in the new version:

vyacht-output-1.3.28.txt

vyacht-output-1.3.30.txt

Other parts of the router software are not. I will upload them as tgz to start with.

Sounds good to me! Anything to help put together my own firmware would be really great.

Also, where is the best place to submit other issues? This repo or some other one (perhaps the new one that pulls everything together)? I am having troubles reconnecting instrumentpanel and also when trying to use more than one app from the iPad. I think there might be a way to make this better using SO_REUSEADDR for the server socket so it can be reconnected even if it is in a timeout state.

[cwensley]

Interesting.. I'm actually not getting the same problem as before with 1.3.30 where it'd update only every 5 seconds, but the wind direction is still not registering correctly.

[benoitg]

@vyacht, @cwensley : I just received my vyacht, and I am definitely interested in building the firmware from source. And I would definitely help @cwensley getting some sort of modern version control on the build system. I'm definitely interested in getting the tarballs to get something started.

[benoitg]

@vyacht Bumping now that I have my vyacht installed, I would REALLY like to play with the source code. Specifically, the fact that Seatalk Wind speed and angle are converted to separate NMEA MVW sentenses seem to confuse other software greatly (even if is seems to follow the letter of the NMEA standard)

So, any chance you can send a list of shell command to reproduce your build environment from scratch? I'm sure I could take it from there.

[cwensley]

@benoitg glad you have it installed!

I've been using mine for a few years now and while it (still) works well, there is so much more it could do.

I'd love to be able to contribute to this project, but sadly it is difficult to do that without the ability to build firmware from source code.

[benoitg]

@cwensley Not having had any answer, I started porting everything for which we have sources to OpenWrt 18.06. The whole thing is at https://github.com/benoitg/vyacht-firmware on branch openwrt-18.06_vyacht

I cherry-picked all the vyacht stuff, so now it share's common ancestry with the openwrt master git, which should make things much easier in the future.

Is jumped straight to 18.06 because:

1. building 15.05 is painful with modern linux
2. the https://github.com/vyacht/ow1505 repository doesn't seem up to date anyway (opened an issue here: Git repo out of date? ow1505#2)
3. carambola2 is still supported in 18.06

Current state:

• It pulls in all the software of the vyacht firmware that we have source code for automatically.
• Modifies said software's config or source so that it builds with the newer OpenWRT.
• It actually builds a firmware successfully (haven't yet tried to run it)

git clone git@github.com:benoitg/vyacht-firmware.git
cd vyacht-firmware
git checkout openwrt-18.06_vyacht
./scripts/feeds update -a
./scripts/feeds install -a
cp diffconfig .config
make defconfig
make download
make -j4

(Doc is here: https://openwrt.org/docs/guide-developer/build-system/use-buildsystem)

Now, this all builds, but my vyacht is at the boat, so I didn't check that the image runs. I'll no doubt have to update some software config to make it actually work. That off course assumes that https://github.com/vyacht/ow1505 isn't missing anything important. Hopefully @vyacht will chip in.

[PeterFromSweden]

@vyacht @cwensley @benoitg
I am struggling to get NMEA-2000 to work with vyacht but have no forum to put it...
I have Raymarine EVO-100 autopilot with Seatalk-ng hooked up to vyacht. Instruments run Nexus and is hooked up together with AIS and GPS to vyacht over NMEA 0183.

I don't know if anything is received over N2K, how to know?

What I need is:

1. Wind data (AWA) from NMEA 0183 to N2K for use with EVO-100
2. Compass data from EVO-100 and N2K for use with WiFi nmea.

I also think that it would be benificial to move packages to openwrt (I've made some contribution upstreaming to mainline openwrt).

Regards
/Peter

[benoitg]

@PeterFromSweden I don't think it's worth sending to openwrt untill we know that it works. So far we don't even know if we have every change that @vyacht made.

Well, my vyacht is at the boat, and it's the middle of winter. If it superficially works for you, I can make another update to latest openwrt. Beyond that, there is not much more I can do right now.

For your information, mine does write to N2K if I enable the checkbox. But it eventually chokes and stops working and needs to be restarted.

[PeterFromSweden]

Ok @benoitg. I was thinking that a redesign of implementation would enable mainstream gpsd. If the spi data was moved into a virtual com-port the gpsd could use these natively with no knowledge of vyacht spi-ports.
By the way, it is winter here too and for me that means that I have time to struggle with vyacht instead of focus on sailing...
Thank you for sharing.

[PeterFromSweden]

@vyacht @cwensley @benoitg.
I have not any connection to N2K (as far as I know). Log file seems to receive only from NMEA0183 (which I know works fine).

A basic wiring question:
http://vyacht.net/opencart/documents/NMEA-WiFi-vyacht-manual-v97.pdf (pg 8)
N2K is here CANL, CANH, Shield (from left to right)

Printed on vYacht is Hi,Lo,Sh (also from left to right...)

Confusion: Where does Hi and Lo go!?!?

Regards
/Peter

gpsd:INFO: vyspi_write: n2kl\x03\x04\xf9\x01 (vyspi:///dev/ttyATH0) ports= 4
gpsd:IO: => GPS: 7e00116e326b6c0304f9010076ff010011f80100 (20, 0)

gpsd:IO: VYSPI reading from device with status 132
gpsd:IO: Read 132 chars to buffer offset 0 (total 132): 7e01162449495657522c2c522c2c4e2c2c4d2c2c2a30320d0a7e01162449495657542c2c522c2c4e2c2c4d2c2c2a30340d0a7e010e2449494844542c2c542a30430d0a7e01112447505743562c2c4e2c2c4e2a35350d0a7e01142447505854452c412c412c2c522c4e2a36450d0a7e01132447505a44412c2c2c2c2c30302c2a34380d0a
gpsd:IO: <= GPS: $IIVWR,,R,,N,,M,,*02\x0d\x0a
gpsd:IO:                awa= nan, aws= nan m/s

[benoitg]

Hi and Lo are from the DeviceNet standard. Normally white is Hi, Blue is Lo, and the shield is bare. But it can depend on who actually manufactured the cable.

[PeterFromSweden]

I knew that, but the confusion is documentation says that hi is where lo is marked on device. I use white as hi marked on device, but what if marking is wrong and documentation is right... I could simply just test to swap I assume. Den ons 29 jan. 2020 kl 17:17 skrev Benoit Grégoire < notifications@github.com>:

…

Hi and Lo are from the DeviceNet standard. Normally white is Hi, Blue is Lo, and the shield is bare. But it can depend on who actually manufactured the cable. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1?email_source=notifications&email_token=ABML7TWZCHIBCYEI7UZXQZ3RAGTY7A5CNFSM4ESFVY52YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKHZNEI#issuecomment-579835537>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABML7TUX2UYYZ2TC5NU7WYLRAGTY7ANCNFSM4ESFVY5Q> .

[benoitg]

I've had problems fixed by reversing polarity, but I can't remember if it was incorrect labeling or a mistake on my part, nor on which port.

[PeterFromSweden]

It was wrongly labeled, now it works. To "fix" problem with gpsd dying after 10-20 min I changed init.d to respawn. Regards /Peter Den tors 30 jan. 2020 kl 06:14 skrev Benoit Grégoire < notifications@github.com>:

…

I've had problems fixed by reversing polarity, but I can't remember if it was incorrect labeling or a mistake on my part, nor on which port. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1?email_source=notifications&email_token=ABML7TT6MVNQAPSF2CB4323RAJO4FA5CNFSM4ESFVY52YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKJWYTI#issuecomment-580086861>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABML7TRH2ZVMS2AD27HCG3TRAJO4FANCNFSM4ESFVY5Q> .