So, re-installing 4.2 fixed everything where I could work on a Panda. However, I’ve been unable to deploy to a Hydra ever since. I’ve uninstalled everything and reinstalled everything listed in the first post. Then if I go create a new Gadgeteer (4.2) project with Hydra and try to run it right off the first thing that happens is I get an error that
because there is a…
using Gadgeteer.Networking;
in Program.cs but there is no matching reference in the template. No problem. Don’t need it anyway (but it would be nice to have the template fixed).
So, I remove the using and run and I get the following output…
I’ve replaced all assembly references multiple times and it is set to use the 4.2 framework. The problem appears to be in the GHIElectronics.Gadgeteer.FEZHydra assembly. I’ve confirmed I’m using the one in the 4.2 directory. Any other ideas?
@ Ian
I’m running some test apps on the Hydra with 4.2 without problems.
I disinstalled all the files (netmf and ghi sdk) from previous 4.1 and reinstalled again from scratch new 4.2 installers. Than I create a new app with template and it works fine, I can deploy and debug. Porting 4.1 projects is a mess. You need to clear all references and delete all dll in the debug dir of the project.
I’ve not yet tested complex apps or particular board. I was just writing some code for test with extender and a 74hc595 shift reg.
I’ve no BSOD on my test PC (AMD E350 cpu)
I wasn’t having any problems running 4.2 on the Hydra until I uninstalled and re-installed to get 4.1 back on for the Panda. Now I can’t get the Hydra to do anything. I’m going to try a different Panda when I get home just to be sure something isn’t messed up on it but I think something isn’t getting installed correctly with 4.2 in this situation.
Any other ideas? I’ve got an IO60P16 driver to finish up
If I can’t get passed this tonight then I’m going to try backing up to 4.1 just so I can get something done.
The netmf team keep trying to close the issue as they cannot reproduce it. They seem to be the only people on the planet who cannot reproduce it.
I have to assume that they do not have the resources to allocate to it. According to the issue log, they currently believe it is specific to Netduino.
Gus, you may have better contact with the netmf team than just the codeplex tracker so they may in fact be better informed than the tracker shows.
For me the BSOD’s were so disruptive that I’ve had to shelve product development using netmf on both GHI and SecretLabs hardware.
Happy to return once it’s fixed.
Woo! Hoo! After uninstalling everything GHI and everything NETMF again, deleting anything left in the directories after uninstall, re-downloading the installs for both again, and reinstalling everything again I can now run a Hello World on Hydra. Four nights lost. Love livin’ on the cutting edge
Sorry, I’m not quite sure I understand what you’re trying to say… If you’re saying that you’ve installed 4.2 but you’re still using 4.1 on the Hydra then I’m not sure what you hope to accomplish.
Ok. Those are the same steps I took. But for some reason I had to take them several times before everything got setup correctly. I didn’t install 4.1 by itself. It was installed with the 4.2 install.
@ ian
Have you tried to reinstall/repair “GHI. NET Gadgeteer SDK v4.2.msi” ? This let me work with 4.1 and 4.2 on Hydra again.
I’ve checked also 4.2 with Hydra too see if I get same problems described here by many people, but I never get a BSOD on win7-x64. The only problem I face is that sometime I can’t deploy on mainboard from VS2010 nor from MFDeploy. In this case I need to “disable” driver and reenable again, and it’s working fine.
Try to press the reset button of your Hydra or Cerberus board or unplug the USB cable while your deploy does not work (freeze). Every time a BSOD occur for me with this scenario. I must stop the deploy process the avoid a crash.
No, I confirm that here it doesn’t crash in BSOD. I see the ghi USB debug driver locked up, but disabling and reenabling start to work again. Tomorrow I will receive a Cerberus board and I will try again
I am able to reproduce a hang and/or BSOD with Cerburus at will. I have found it to be less tempormental with USB cable plugged directly into my laptop instead of through a hub (or two) in case that helps anyone.
Ok, I finally got a FEZ Cerberus board. Flashed with 4.2 and start deploying very basic led blinking program. No problem at first time, program running fine. Deploying the second time, I got not ping, VS2010 nailed. Pressing reset on Cerberus, I got BSOD !! Retrying on Hydra, no BSOD at all.
Back to Cerby and no way to deploy again. I noted that in the MFDeploy I got a warning relative a delay more that 10s. After few seconds from reset the board become unusable and unpingable, but I’ve no tight loop, no timers, only a thread running with a 250ms sleep inside.
Ok, reset Cerby and immidiately deploy from VS, finally it seems unlocked and working fine. Probably when deploying a program if the board get stuck, something wrong happens on CLR on the board that become unresponsive. The trick I find is fast deploy from VS immidately after reset, in this way I avoid to flash again fw any time Cerby locks.
