Day 1: I only have to touch things and they break ...

My first binary (test1.bin) didn't drag/drop into mbed properly, after a while I got a system message. Anyway ... now I have a persistent file. I can Delete it, and it appers to go, but when I Refresh, its back. Perversely I can re-name it! I can download other files and delete those ... hardly a major issue, but any ideas.

That happened to me as well when using the mbed on a Windows 7 PC.  Then the problem mysteriously went away by itself.

Hi Ian, Alessandro,

Can you give us any more specs on your setup? It maybe that windows 7 has added some "features" that haven't shown themselves in testing that we should investigate.

Thanks,
Simon

Hello Ian and Alessandro.

That problem seem to suggest that the Mbed filesystem index has some how got corructed. It can some time happen when a drive loses communication or power in the middle of task like copiying. i've had similar problem in the past with USB drive when I disconnected the drive to early. The quick solution, if you don't care for the lost data,  is to reformat your Mbed dirve.

hope that helps and good luck.

Ronald Sousa

Mines XP ... I was running via USB expander at the time.

When I initialy tried to delete it it appeared to go (But didn't of course). So when I tried to write a new file in, with the same name ... it mutters domething about corruption. And it wasn't obvious that it hadn't sucessfuly writen the file, because befoe/after file name was the same.

I can try a reformat of the EMBED drive ... Guess I'll need to backup the MBED.HTM before I do (Maybe its not necessary if it is in ROM?)

Cheers.ian

Hi Ian,

I can try a reformat of the EMBED drive ... Guess I'll need to backup the MBED.HTM before I do (Maybe its not necessary if it is in ROM?)

If you delete MBED.HTM, it will appear again after a powercycle. Its the file that wont quit!

Thanks,
Chris

Simon Ford wrote:

Hi Ian, Alessandro,

Can you give us any more specs on your setup? It maybe that windows 7 has added some "features" that haven't shown themselves in testing that we should investigate.

Thanks,
Simon

Simon,

It's a Windows 7 Ultimate system with an Intel Core i5 and Asus P7P55D motherboard.  It's also got Norton Internet Security 2010.  Another peculiar behavior it had was that I could not overwrite any files as it would almost certainly hang internet explorer as it tried to write the file.  I would then have to brutally remove the USB cable to get out of that situation.  However, sometimes it would hang even when writing a new file.

Hiya Alessandro,

By any chance, does your program use the LocalFileSystem?

Ronald Sousa,

Ronald,

I'm not entirely certain that the specific program I was trying to load used LocalFileSystem.  However I also remember now that windows reported a much higher used space than the sum of the file sizes on the disk.  Hence I'm leaning towards the possibility that the filesystem got corrupt somehow, as you mentioned before.

When I get a chance, I will try to reproduce the issue more reliably.

Alessandro

Well I had a go at (Re)Formatting the device. It came up with the usual dialogue, progress bar moved across ... then long pause. After about 2min message says 'Windows was not able to complete this action'. BUT, when I re-mounted the drive the troublesome file has gone ... and the MBED.HTM has re-generated (As Chris said).

I guess the Windows message represents a puzzled system, which having formatted the drive expected to find it empty!

Hi Ian,

Thanks for the update.

I'm not an expert this area, but I know that that MBED.HTM is written to the disk on power up if the firmware finds the drive empty. I'd expect that after a format, the disk really would be empty until the next power cycle when the MBDE.HTM file is created.

Just to rule out any generic problems, can you sucessfully reformat a standard USB flash disk on your Windows 7 PC?

Just to be clear, you said that after the reformat the troublesome file had gone, does this mean it is now behaving as you'd expected?

Thanks,
Chris

Thanks,
Chris