3.2 crashes repeatedly

[bobofthenorth]

I dunno anything about Apples, but 3.2.0 for IntelMac is available for download here.

[transformer]

next time someone can try to connect toaster or vibrator to OCPN and see does it crash. it is free, no harm done.


and I could not edit my previous post. I noticed few mistakes.

instead word "useful" should be "useless".

[Tar34]

Originally Posted by bobofthenorth

I dunno anything about Apples, but 3.2.0 for IntelMac is available for download here.

If you enlarge the photo in my post #15, you will see 3.2.0 is what I have installed. What I experience is what you see. Dimmed flat display without access to function buttons.........

[bobofthenorth]

Tar34 you need to start a new thread with a descriptive subject that will attract attention to your problem. Posting inside this thread is not going to be helpful for you. I've already said that I don't know anything about Apples.

And Transformer I fear you may have a short in your windings. Perhaps you can find someone to check that out for you.

[HappySeagull]

Originally Posted by bobofthenorth

The plot thickens .......

Last night I realized that when I'm underway in addition to the autopilot and O running I also have my radio turned on. It didn't seem to me that it could possibly be related but my radio does receive the GPS coordinates from the autopilot so last night I left the radio on. When I got up this morning O had crashed. I've got it restarted now and I'll watch for the flashing position information but in the case of the crash last night it occurred without any zoom or pan because I was asleep when it happened.

So just to be clear:
- O runs on a W7 64 bit laptop with a USB GPS puck plugged into it.
- O is outputting to my autopilot on COM1. COM1 is a 2-way port so that O can receive mag heading information from my autopilot
- my radio is connected directly to my autopilot on one of the autopilot's output ports.

THAT's interesting- I haven't got that kind of gear so can't follow along very well but keep stirring the pot!

[dynamo]

I did find that I could cause 3.2 to crash repeatedly after runnning for a while when I fed it with corrupted NMEA data. The reason for this corruption was that I had accidentally used an isolated RS422/485 differentail USB to serial converter on a single-ended feed from my Raymarine plotter. Many NMEA sentences got through ok, but some were corrupted. I didn't flag this as a problem at the time because OCPN worked fine after I changed the USB to serial converter, my mistake. Perhaps you should log your NMEA data & check for corruption?

[bobofthenorth]

I suppose it is possible that somehow the NMEA stream is getting corrupted by feeding the radio but I can't see how. The radio is an NMEA listener and its fed from an out only port on the autopilot so I'm baffled as to how it could be the problem. But OTOH I've had O running for over 24 hours now with the radio off and still no crashes. I'm going to let it run a full 48 hours just to make sure that the first time wasn't coincidentally time related when I turned the radio on.

Got a reccomendation for an NMEA data logger? And how would I recognize a corrupt stream?

[wdkester]

Originally Posted by bobofthenorth

I suppose it is possible that somehow the NMEA stream is getting corrupted by feeding the radio but I can't see how. The radio is an NMEA listener and its fed from an out only port on the autopilot so I'm baffled as to how it could be the problem. But OTOH I've had O running for over 24 hours now with the radio off and still no crashes. I'm going to let it run a full 48 hours just to make sure that the first time wasn't coincidentally time related when I turned the radio on.

Got a reccomendation for an NMEA data logger? And how would I recognize a corrupt stream?

I don't understand the utility of having a "radio" as a NMEA listener. What is it listening for?

[yachtman]

Originally Posted by wdkester

I don't understand the utility of having a "radio" as a NMEA listener. What is it listening for?

position sentence used with DSC radios, so it can transmit position and MMSI number all at the press of a button, for emergency calls etc

[bobofthenorth]

Evidently the radio was a red herring. After about 40 hours running with the radio turned off I got a crash last night. So the radio listening appears to shorten the time to a crash but the crash happens regardless. And I think its safe to say the crash happens much sooner if the boat is underway. Unfortunately I'm going to be off the boat for about a month now so I won't be able to try anything else although I think I'm about at the limit of my troubleshooting abilities anyway.

[LH44 Anne]

I've been using Fugawi for some time, using both raster and vector charts from NOAA (U.S. East and Gulf coasts) and raster charts from Nautical Publications NV (Bahamas, Puerto Rico, Turks & Caicos, Dominican Republic). No problems. (Also have Raymarine chart plotter with similar coverage with Navionics cartridges.) No problems.

I've recently installed O CPN and "loaded" or linked to the existing charts on my machine. And I've prowled around on the forum a good bit, and have learned a good bit. In fact, it is the forum which has attracted me to O CPN in the first place.

I'm running O CPN 3.0.2 on a 32 bit Win 7 Pro, SP1, 2.53 Gb Core i3 Intel CPU, and 6 Gb RAM. All updates are current and machine is normally quite stable and problem free. (Including running Fugawi with the installed charts.)

While the system is my "boat computer," I'm currently running it at home, not on-board. So far, I've always run it as a stand alone, with no other applications running. Only security software and a few "TSR" utilities (to apply an archaic term) and basic OR routines are running processes. I've vigorously weeded out all the crapware that came with the system and shut off the "start at boot" routines for almost all applications.

My USB GPS is not installed at present.

Now the issues: O CPN loads and runs - apparently normally - and then - suddenly, with no warning - closes (crashes?). It occurs anytime from ten minutes (occasionally less) after load to as much as 45 minutes (occasionally an hour, but not very often). There is no error message from O CPN, but Win 7 pops up an "abnormal closure" routine, offers to find a fix, never does, and then closes itself and O CPN. So far as I can see, it does not interfere with the system stability or leave garbage in RAM. The surest "trigger" is to visit a chart that has not been opened previously, which begins the file conversion process but the "closure" seems to occur at unpredictable occasions even when nothing active is going on.

Don't know what other information I can provide that would be relevant. Any suggestions? Any answers? Pretty please?

[bdbcat]

LH44...

You may want to try the newest Version 3.2. Many improvements in stability, especially on Win7.

Thanks
Dave

[tiersp]

Hello,
is the option control the checksum clicked when it crashe ?

[bobofthenorth]

Yes, control checksum is checked for both streams.

[LH44 Anne]

Originally Posted by bdbcat

LH44...

You may want to try the newest Version 3.2. Many improvements in stability, especially on Win7.

Thanks
Dave

Outstanding improvement. Based on short time since update to 3.2, no crashes (12 hours and counting, compared to ~10 minutes before). Much faster performance. Much better chart handling as well. At this point, you're giving my Fugawi installation a solid run for "best on my machine" honors. I'm looking forward to trying (and comparing) on the water.

Haven't yet had time to explore everything, but so far, 3.2 is a VERY BIG improvement.

Thanks, Dave!