is that ai super jumbo image healthy?

GoogleGroups
1

i've tried twice to download that image, and both times, i've ended
up with a download that seems corrupted and whose md5sum doesn't match
what's listed here:

  http://www.alwaysinnovating.com/beagleboard/

i've tried to download the image from amazon, and this is what i end
up with:

-rw-r--r-- 1 rpjday rpjday 2191575252 2011-04-04 17:13 ai-superjumbo-2011-03.a.gz

but any attempt to gunzip tells me

"gzip: ai-superjumbo-2011-03.a.gz: invalid compressed data--format
violated"

  i know others were getting this a few days ago.

rday

p.s. and, no, the md5sum for the gz file isn't correct.

2

I ended up buying a one month premium membership to get the big file the first day it appeared. It was worth it to me and worked the first time, although I need to resize it to make better use of an 8gb card. I’ d rather build it next time so updating would be easier.

Mark

3

that still doesn't address the issue that the image at amazon
appears to be borked. i've tried three different times, and gotten an
allegely corrupted image all three times.

rday

4

The only way I could get a good image was to download the 3 bits and paste
them together. The torrent that Jason put up was also broken.

5

ok, thanks for confirming that. over to you, jason. :slight_smile:

rday

6

i've tried twice to download that image, and both times, i've ended
up with a download that seems corrupted and whose md5sum doesn't match
what's listed here:

Always Innovating: Introducing the MeCam

i've tried to download the image from amazon, and this is what i end
up with:

-rw-r--r-- 1 rpjday rpjday 2191575252 2011-04-04 17:13 ai-superjumbo-2011-03.a.gz

What type of file system are you downloading to? This almost sounds
like the FAT 2GB barrier, except that that file size is the same that
I downloaded.

but any attempt to gunzip tells me

I was able to gunzip my original download, but I haven't been able to
test the image yet because my 8GB SD cards are smaller than the 8GB
image. :frowning:

My torrent downloader is showing me problems, but the .gz downloads
fine and gives me an md5sum of 4352bc1a2e474504b7ae74c314788bdc. Sure
enough, that doesn't match the md5sum on the AI website:
3ab5edc44bb73f3417f67fa070857157.

The download link is too slow for me to practically even try again.
If someone can give me a copy of the right data, I will again try to
share to provide bandwidth.

7

I ended up buying a one month premium membership to get the big file the
first day it appeared. It was worth it to me and worked the first time,
although I need to resize it to make better use of an 8gb card. I' d rather
build it next time so updating would be easier.

I'll just go ahead and do the same.

8

downloading to my ubuntu 10.10 system -- 64-bit machine, ext4
filesystem. since i recall older browsers having a download limit, i
downloaded using both recent chrome and firefox -- same result each
time.

rday

p.s. here's my md5sum:

$ md5sum ai-superjumbo-2011-03.a.gz
4352bc1a2e474504b7ae74c314788bdc ai-superjumbo-2011-03.a.gz
$

  if others are getting precisely the same md5, that's pretty
conclusive that it's the image.

9

I just did a new download of the .tbz2 file and I get the same md5sum
as listed on the AI website. I have a slow DSL uplink, so it will
take some time to put up on S3, but I am doing that now. I think AI
must have changed the image at some point since I originally
downloaded, since even the file extension has changed.

I just booted this new image and it seems to work. I'll check tonight
when I get to California if the upload completed.

10

Dear Grégoire,

Thank you very much for sharing your innovations with the rest of us!
I am also a beliver of the open SW/HW movement since it is beneficial
for so many people in many ways.

I would now like to try your creation, the Super-Jombo.
I would also appreciate if you could help me to get started in the following
situation:

First, I had problems in trying to download with your download option 3:

−

NoSuchKey
The specified key does not exist.
demos/ai-superjumbo-2011-03.tbz2
9D8A018333037909
−

/K1VC9zK2Er8KmRHrwPmiDk4i66quEeSKR5RUwwblrQScdsikir7jKnnOGbW2KC9

What is wrong?

I then tried option 2 (with platinum $6.99). The download was quite quick but being new to Linux I could not
create the SD card out of the single ai-superjumbo-2011-03.a.tbz2 file.

As my last hope I went to option 1 and was successful both in downloading (13 h) and in
generating the three partition SD card:

root@sin-laptop:/media/SUPER-JUMBO# gunzip -dc /home/sin/Downloads/ai-superjumbo-2011-03.a.gz | dd of=/dev/mmcblk0 bs=1M
dd: writing `/dev/mmcblk0’: No space left on device
0+116305 records in
0+116304 records out
3980394496 bytes (4.0 GB) copied, 559.84 s, 7.1 MB/s

root@sin-laptop:/media/SUPER-JUMBO# fdisk -l

Disk /dev/sda: 80.0 GB, 80026361856 bytes
255 heads, 63 sectors/track, 9729 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x0002934d

Device Boot Start End Blocks Id System
/dev/sda1 * 1 9356 75143168 83 Linux
/dev/sda2 9356 9730 3005441 5 Extended
/dev/sda5 9356 9730 3005440 82 Linux swap / Solaris

Disk /dev/mmcblk0: 3965 MB, 3965190144 bytes
255 heads, 63 sectors/track, 482 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/mmcblk0p1 * 1 93 746991 c W95 FAT32 (LBA)
/dev/mmcblk0p2 94 411 2554335 83 Linux
/dev/mmcblk0p3 412 473 498015 82 Linux swap / Solaris
root@sin-laptop:/media/SUPER-JUMBO#

Now my problem is that my BeagleBoard C4 does not even boot with this
new SD card.

Please advice about what I should do!

Best regards,
Seppo Nikkilä
Helsinki, Finland
(currently Estepona, Spain)

11

Here is the completed upload:
http://beagle.s3.amazonaws.com/demos/ai-superjumbo-2011-03.a.tbz2

Note that this one really is a tarball and you have to extract the
uncompressed SD card image out of it.

12

Here is the completed upload:
http://beagle.s3.amazonaws.com/demos/ai-superjumbo-2011-03.a.tbz2

  so just to be clear, that is *not* the same file as the one
corresponding to the link at the AI page here:

since *that* one is a gz, not a tbz2 file, correct? so the link on
the AI page will still give you a corrupted file.

Note that this one really is a tarball and you have to extract the
uncompressed SD card image out of it.

  got a checksum for this one? i'm downloading as we speak -- should
be done in a few minutes.

rday

13

Hi,

Now my problem is that my BeagleBoard C4 does not even boot with this
new SD card.

Please advice about what I should do!

you card is 4GB, the image fits on a 8GB
you can manually copy the content of the second partition this way :
mount -o loop,offset=764951040 ai-superjumbo-2011-03.a /mnt/
fdisk /dev/mmcblk0
-> create the 2nd partition with the right size for your SDCard
mkfs.ext3 /dev/mmcblk0p2
mount -t ext3 /dev/mmcblk0p2 /mnt2
cd /mnt
rsync -avz . /mnt2/
umount /mnt2

Eric

14

Thank you Eric,

You made my day!

Gratefully,
siñ

15

just FYI, here's what i get from that downloaded file:

$ md5sum ai-superjumbo-2011-03.a.tbz2
3ab5edc44bb73f3417f67fa070857157 ai-superjumbo-2011-03.a.tbz2
$ md5sum ai-superjumbo-2011-03.a
f6c1fa74f274fbe8f117e9a884a086e6 ai-superjumbo-2011-03.a
$ file ai-superjumbo-2011-03.a
ai-superjumbo-2011-03.a: x86 boot sector; partition 1: ID=0xc, active,
starthead 1, startsector 63, 1493982 sectors; partition 2: ID=0x83,
starthead 0, startsector 1494045, 5108670 sectors; partition 3:
ID=0x82, starthead 0, startsector 6602715, 996030 sectors, code offset
0x0
$

  so just those quick tests look good. onward ...

rday

16

Here is the completed upload:
http://beagle.s3.amazonaws.com/demos/ai-superjumbo-2011-03.a.tbz2

so just to be clear, that is *not* the same file as the one
corresponding to the link at the AI page here:

Always Innovating: Introducing the MeCam

since *that* one is a gz, not a tbz2 file, correct? so the link on
the AI page will still give you a corrupted file.

Correct.

Note that this one really is a tarball and you have to extract the
uncompressed SD card image out of it.

got a checksum for this one? i'm downloading as we speak -- should
be done in a few minutes.

The new checksum matches what is on the AI page.

17

I got:

root@sin-laptop:/media/SUPER-JUMBO# md5sum /media/SUPER-JUMBO/ai-superjumbo-2011-03.a.tbz2
3ab5edc44bb73f3417f67fa070857157 /media/SUPER-JUMBO/ai-superjumbo-2011-03.a.tbz2
root@sin-laptop:/media/SUPER-JUMBO# md5sum /media/SUPER-JUMBO/ai-superjumbo-2011-03.a
6ca5e3920bb3b38b880fb2df04fdfb33 /media/SUPER-JUMBO/ai-superjumbo-2011-03.a
root@sin-laptop:/media/SUPER-JUMBO#

So the .tbz2 file is right but the .a file is wrong.

What are the steps to generate the working SD card from the .tbz2 file and are the old partitions
overwritten automatically on the card? What size card 4GB or 8GB should I use.
Does the Super-Jumbo run on the standard Beagleboard C4?

siñ

18

that's weird -- i got the same (correct) md5sum for the tbz2 file,
but i got:

$ md5sum ai*
f6c1fa74f274fbe8f117e9a884a086e6 ai-superjumbo-2011-03.a
$

and as i reported earlier (but haven't had a chance to boot from yet),
the file command suggests i have a perfectly reasonable image:

$ file ai-superjumbo-2011-03.a
ai-superjumbo-2011-03.a: x86 boot sector; partition 1: ID=0xc, active,
starthead 1, startsector 63, 1493982 sectors; partition 2: ID=0x83,
starthead 0, startsector 1494045, 5108670 sectors; partition 3:
ID=0x82, starthead 0, startsector 6602715, 996030 sectors, code offset
0x0

  can someone verify that that's what *they* get from the "file"
command? sure looks like a properly partitioned image to me.

rday

19

My data are the same, with the file extracted from the 3 separate pieces.

20

Robert P. J. Day wrote:

I got:

root@sin-laptop:/media/SUPER-JUMBO# md5sum
/media/SUPER-JUMBO/ai-superjumbo-2011-03.a.tbz2
3ab5edc44bb73f3417f67fa070857157 /media/SUPER-JUMBO/ai-superjumbo-2011-03.a.tbz2
root@sin-laptop:/media/SUPER-JUMBO# md5sum /media/SUPER-JUMBO/ai-superjumbo-2011-03.a
6ca5e3920bb3b38b880fb2df04fdfb33 /media/SUPER-JUMBO/ai-superjumbo-2011-03.a
root@sin-laptop:/media/SUPER-JUMBO#

So the .tbz2 file is right but the .a file is wrong.

   that's weird -- i got the same (correct) md5sum for the tbz2 file,
but i got:

$ md5sum ai*
f6c1fa74f274fbe8f117e9a884a086e6 ai-superjumbo-2011-03.a
$

and as i reported earlier (but haven't had a chance to boot from yet),
the file command suggests i have a perfectly reasonable image:

$ file ai-superjumbo-2011-03.a
ai-superjumbo-2011-03.a: x86 boot sector;
partition 1: ID=0xc, active, starthead 1, startsector 63, 1493982 sectors;
partition 2: ID=0x83, starthead 0, startsector 1494045, 5108670 sectors;
partition 3: ID=0x82, starthead 0, startsector 6602715, 996030 sectors,
code offset 0x0

   can someone verify that that's what *they* get from the "file"
command? sure looks like a properly partitioned image to me.

md5sum ai-superjumbo-2011-03.a

1ce14963c18d728bd2cbdc329f3b97a8 ai-superjumbo-2011-03.a

file ai-superjumbo-2011-03.a

ai-superjumbo-2011-03.a: x86 boot sector;
partition 1: ID=0xc, active, starthead 1, startsector 63, 1493982 sectors;
partition 2: ID=0x83, starthead 0, startsector 1494045, 5108670 sectors;
partition 3: ID=0x82, starthead 0, startsector 6602715, 996030 sectors