cRio not working...?

[Jack Gillespie]

We're trying to deploy the labview code to the cRio. However, the cRio (10.xx.yy.02 and .03 ) ip addresses/devices are not showing up, which (WE THINK) is why we can't download/deploy anything to the robot. We can see .01 (the bridge), .04 (the ds wireless), .05 (the ds), and .06 (the computer).

My question is, could the cRio somehow be broken/fried and how would we tell? It still has the little green lights that light up when the ethernet cord for the bridge is plugged in.

[Greg McKaskle]

If you no longer get lights of any sort to come on on your cRIO, then will have a chance to send it back in.

Greg Mckaskle

[crake]

Actually you have power if the ethernet lights are on. Is the status light blinking?

[Jack Gillespie]

The little green lights on the red cards are all on. the green light next to the word "power" is on. when the robot is first turned on, the yellow light next to the word "status" lights up, then turns off.
But it doesn't blink or reappear again.
Still confused, though. (See, I am so not a programmer here!)

[crake]

The little green lights on the red cards are all on. the green light next to the word "power" is on. when the robot is first turned on, the yellow light next to the word "status" lights up, then turns off.
But it doesn't blink or reappear again.
Still confused, though. (See, I am so not a programmer here!)

Looks like the cRIO successfully booted and is ready. I know other teams have had similar configuration issues - let me see if I can find some posts...

[Jack Gillespie]

The only thing is, we've never gotten to deploy the actual labview robot project code. Everytime it says "trying to connect to the cRio" message and doesn't get through. That's why we looked at the linksys manager and found out that the .2 and .3 aren't showing up. We freaked out that we had somehow fried it. So if it's all booted and ready and stuff, I'm confused why it doesn't ever say it was deployed successfully....

Thank you, though! You've made me much much less anxious that we had somehow killed it.

[crake]

Here's a similar thread http://www.chiefdelphi.com/forums/showthread.php?t=73620

[Jack Gillespie]

Ok, read through that. I'm not sure what all that means, though. That had a .2 and they were trying to update? We don't have the .2, and I guess we're trying to update too. But honestly, I'm clueless. I can get the crio hooked up directly (like the thread said), but it still won't show a .2 or a .3. I'll go try again, though.

Thanks again - I printed out that thread to take back to the other room (where the robot is doesn't have a wifi for internet, only for the crio. sucks!)

be back shortly!!

[Jack Gillespie]

ok, here i go again.

I tried hardline connecting ds-laptop-crio like the other thread mentioned. Even when plugged directly into the crio, I did not get the .2. Meaning that the computer/system is not reading the crio at all.

Now I'm completely confused, because it was directly connected to it!!

[crake]

ok, here i go again.

I tried hardline connecting ds-laptop-crio like the other thread mentioned. Even when plugged directly into the crio, I did not get the .2. Meaning that the computer/system is not reading the crio at all.

Now I'm completely confused, because it was directly connected to it!!

Assuming you have a good link (Enet lights are lit and blinking), can you ping the controller at all? Any reason you might have changed the IP address? Have you read the instructions in the previous link or tried the network address changes?

[Jack Gillespie]

Assuming you have a good link (Enet lights are lit and blinking), can you ping the controller at all?
...um, ok, will run up there again and ask the other teacher to do that (I'm the math side, he's the electronics side). But I don't have a clue what that means, personally.

Any reason you might have changed the IP address?
Nope, and there's only one IP that we can easily change (the laptop), which isn't the problem.

Have you read the instructions in the previous link or tried the network address changes?
Read the instructions, gave to jack (the other teacher) to read and make sure we had tried that, but I think the changing of the laptop isn't the problem - it's getting the cRio to respond, and we didn't see that happen when we changed the ip earlier.

But will try again after printing out this thread too, so jack can check that I have done the things you and the other thread suggested.

be back shortly! Fingers crossed....

[crake]

In the morning, if you're still stuck, please visit ni.com/first and we can help.

[Jack Gillespie]

well, we pinged it, and it responded. Very pleased!

Here's what Jack and I realized might be the issue - our DS is saying"Battery: 00.00V" when we've tested the battery on the robot and it is running 12V (actually, when we checked, 12.96 or something.) Wondering if that could be the issue and it's why we can't see/talk to the crio?

[crake]

That's a good sign! But I'm pretty sure it isn't the battery voltage. If the lights are on, and you can ping the device then the cRIO is up and running. 12V battery output should result in a healthy 24V going to the controller. It seems like an imaging problem, and you should be able to follow the imaging steps and get up and running from this point.

[Jack Gillespie]

Wow! You were so helpful last night. Now we can deploy all of it, but 2 things crop up.
1. During build, it says it prevented a "file name error" or "duplicate file error" or something. We think it means 2 files are named to same thing? Have you any clue how we could find out WHICH file is the problem?
2. during deployment, it says that it failed to deploy/load startuplibrarypatch.out (i may be missing a space, it was last night). Joysticks and auto mode still do nothing, however.

Any ideas? I have the kids searching the computer/project for the file, but otherwise I'm still stymied, because if it deployed everything else, shouldn't it work somehow?

THANKS!

[Greg McKaskle]

The first thing is just a warning. It will happen anytime you use WPI because there are multiple VIs names Open for instance.

The second issue has happened before, but I don't know the workaround. If you go into the cRIO imaging tool, it has a checkbox for the startup app, and if you turn it off, I think this will go away. I'm sure there is another way to deal with it, but I don't know off the top of my head.

Greg McKaskle

[Jack Gillespie]

so i guess the question is, does that one file prevent the cRio from running code? Because after its deployed the rest, we try moving the joysticks or putting it into autonomous mode, but nothing happens. dead robot is not a good robot, even if we are the zombots.

[Selijdice]

um...so we re-imaged, and thought we had it fixed - and now the status light is blinking, and mocking us.

uhoh.

what the heck does that mean?

[crake]

What is the blink pattern?

[Jack Gillespie]

Blink twice then off for 1-2 seconds, Repeat pattern

Jack Gillespie

[crake]

This is an internal SW error - and is something I've never actually seen. However I've heard of one other team that had this issue. Let me ping some folks here.

[TDohse]

Have you updated your LabVIEW to 3.0a, and re-imaged with the file FRC_2009_v11.zip?

Do you have a null modem cable available so you can see the console output? If you toggle the "CONSOLE OUT" dip switch to on, then connect a cable from the serial port to your PC you can watch the output in a terminal window (XP provides one for you called HyperTerminal, Start->Programs->Accessories->Communications->HyperTerminal, set the baud rate to 9600). The information printed there while booting has a high likelihood of giving us what we need to diagnose the problem.

Most likely though, if you run the imaging tool again it will fix the problem. The primary design goal was to restore the device to a usable configuration no matter how screwed up the device became.

[Danny Diaz]

The double-blink is a startup library or dependency loading problem; generally when the double-blink occurs, it means that the software stack was unable to properly load. This could be due to missing symbols in a startup library (usually due to out-of-order loading), missing libraries (libraries specified to be loaded that cannot be found), or a bad installation of the software to the target.

When we have a problem like this, the first step in debugging is to enable console out, connect the serial port to a PC via a null modem serial adapter cable, and use Hyperterminal or TerraTerm to capture the serial output of the controller. If there are missing symbols you will get a print to the screen specifying what symbols are missing, and if there is a library loading problem we can usually figure out what went wrong by the software state specified on the console output.

If you are dropped into a "No Software Installed" or "Unconfigured" state (status will be shown on the console out) then that means some .out module listed in the NI-RT.INI file in the root directory of your controller could not be found on the system. You can connect to the cRIO via your favorite FTP client (I highly recommend FileZilla) and view the directory contents on the controller; in the root directory is our primary configuration file NI-RT.INI, and it lists all of the .out files (like target.out, lvrt.out, and so on) that are required for successful loading of the software on your controller. The easiest way to troubleshoot problems with a double-blink of the status LED is to ensure all of the .out files that are found in the NI-RT.INI file - no matter what token or list they are in - are actually on the controller. If you find you are missing a file, you may need to re-image the controller.

The most useful information you can give me to help you from this point would be a dump of the output from the console out during boot of the controller. Then, depending on the output of that dump, checking to make sure all of the .out files listed in the NI-RT.INI file are found in the NI-RT/SYSTEM directory of the controller (or at the path specified if there is a path given) will take you even farther.

I look forward to your response.

-Danny