[rapha]

Hi all,

after Googling around and not finding a solution to my problem I found this forum and happily saw it has a Linux Users Corner

After my old 7730 recently broke (keyboard stopped working, lots of sand and water over the years) I just got another one off eBay which looks and feels like new. I had always thought the USB interface of my old 7730 to be defective, but now with the new one showing the same behaviour when plugging it into the computer the problem must be with Linux (works on a friend's Windows PC with both devices, d'uh!).

What happens is, when I plug in either one of the 7730's and let /var/log/messages scroll by, is this:

Oct 14 13:20:27 momo kernel: [13131.488000] usb 2-1: new full speed USB device using uhci_hcd and address 2
Oct 14 13:20:28 momo kernel: [13131.672000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:28 momo kernel: [13131.748000] usb 2-1: USB disconnect, address 2
Oct 14 13:20:28 momo kernel: [13132.508000] usb 2-1: new full speed USB device using uhci_hcd and address 3
Oct 14 13:20:29 momo kernel: [13132.700000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:29 momo kernel: [13132.760000] usb 2-1: USB disconnect, address 3
Oct 14 13:20:29 momo kernel: [13133.516000] usb 2-1: new full speed USB device using uhci_hcd and address 4
Oct 14 13:20:30 momo kernel: [13133.696000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:30 momo kernel: [13133.768000] usb 2-1: USB disconnect, address 4
Oct 14 13:20:30 momo kernel: [13134.516000] usb 2-1: new full speed USB device using uhci_hcd and address 5
Oct 14 13:20:31 momo kernel: [13134.700000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:31 momo kernel: [13134.704000] usb 2-1: USB disconnect, address 5
Oct 14 13:20:31 momo kernel: [13135.320000] usb 2-1: new full speed USB device using uhci_hcd and address 6
Oct 14 13:20:31 momo kernel: [13135.504000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:31 momo kernel: [13135.544000] usb 2-1: USB disconnect, address 6
Oct 14 13:20:32 momo kernel: [13136.292000] usb 2-1: new full speed USB device using uhci_hcd and address 7
Oct 14 13:20:32 momo kernel: [13136.480000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:32 momo kernel: [13136.484000] usb 2-1: USB disconnect, address 7
Oct 14 13:20:33 momo kernel: [13137.300000] usb 2-1: new full speed USB device using uhci_hcd and address 8
Oct 14 13:20:33 momo kernel: [13137.480000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:33 momo kernel: [13137.484000] usb 2-1: USB disconnect, address 8
Oct 14 13:20:34 momo kernel: [13138.104000] usb 2-1: new full speed USB device using uhci_hcd and address 9
Oct 14 13:20:34 momo kernel: [13138.284000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:34 momo kernel: [13138.316000] usb 2-1: USB disconnect, address 9
Oct 14 13:20:35 momo kernel: [13139.068000] usb 2-1: new full speed USB device using uhci_hcd and address 10
Oct 14 13:20:35 momo kernel: [13139.260000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:35 momo kernel: [13139.264000] usb 2-1: USB disconnect, address 10
Oct 14 13:20:36 momo kernel: [13140.072000] usb 2-1: new full speed USB device using uhci_hcd and address 11
Oct 14 13:20:36 momo kernel: [13140.256000] usb 2-1: configuration #1 chosen from 1 choice
Oct 14 13:20:36 momo kernel: [13140.260000] usb 2-1: USB disconnect, address 11

(and so on, infinitely...)

Right now, I'm on Ubuntu 7.10 (kernel 2.6.22-12-generic) but can remember the same behaviour with the old 7730 on something like Breezy Badger, when I blamed it on the BlackBerry. I have Barry 0.9 running; btool -l comes back empty.

Any ideas?

Regards,
Raphael

[rivviepop]

OK #1 - one of my laptops is named momo too. how funny.

It looks like the problem is because you have Barry installed, and something in the setup and the 7730 interaction is causing the bcharge tool to keep running over and over and over again, resetting the USB port. At least, that's what it looks like from here right now...

What happens if you uninstall all of the barry tools so that no udev rules run, there's no bcharge, etc.? Does the exact same thing happen, or do we get different results? (working or not)

Also try looking in /var/log/dmesg -- certain module dump messages end up in there and not in /var/log/messages, there might be some more logging info that'll help us.

[rapha]

Originally Posted by rivviepop

OK #1 - one of my laptops is named momo too. how funny.

Funny indeed - mine is named after the book Momo by Michael Ende, although I don't know if it is known outside Germany...

Originally Posted by rivviepop

It looks like the problem is because you have Barry installed, and something in the setup and the 7730 interaction is causing the bcharge tool to keep running over and over and over again, resetting the USB port. At least, that's what it looks like from here right now...

I'm not sure if it has been the same before I installed Barry, so I'll try uninstalling it.

Originally Posted by rivviepop

What happens if you uninstall all of the barry tools so that no udev rules run, there's no bcharge, etc.? Does the exact same thing happen, or do we get different results? (working or not)

Also try looking in /var/log/dmesg -- certain module dump messages end up in there and not in /var/log/messages, there might be some more logging info that'll help us.

Okay, /var/log/dmesg doesn't change at all after attaching the BB. /var/log/messages looks exactly the same. modprobe'ing berry_charge doesn't change a thing either (the BB charges fine with or without berry_charge loaded btw, always has). I also tried this on a desktop computer still running Edgy Eft, with both 7730's and same results...

[rivviepop]

Originally Posted by rapha

Funny indeed - mine is named after the book Momo by Michael Ende, although I don't know if it is known outside Germany...

Nope, afraid I don't know that - mine is after the American gangster, Salvatore (Sam) Giancana, aka Momo Giancana.

Okay, /var/log/dmesg doesn't change at all after attaching the BB. /var/log/messages looks exactly the same. modprobe'ing berry_charge doesn't change a thing either (the BB charges fine with or without berry_charge loaded btw, always has). I also tried this on a desktop computer still running Edgy Eft, with both 7730's and same results...

So even without barry installed it's sitting there going over and over and over? Ugh. Can you open another window, go root, and run "lsusb -v >output.txt" while it's plugged in? What we're looking for is the specific block entry for 'Research in Motion' to have a look at the USB details that the computer sees, that might provide a clue to what's going on here... it obviously is too old to have an SD card so it's not USB Mass Storage....

Actually, that's interesting -- what happens if you force remove usb_storage module from the kernel (rmmod usb_storage) and *then* plug in the 7730? (maybe the module *thinks* it's mass storage by accident and is messing up?)

[rapha]

Originally Posted by rivviepop

Nope, afraid I don't know that - mine is after the American gangster, Salvatore (Sam) Giancana, aka Momo Giancana.

Amazon.com: Momo (Puffin Books): Books: Michael Ende ... it's a book for everybody who has to do with time, which, in a 4-dimensional space-time, is pretty much really everybody. Actually it has been written for children, but IMHO it's even more interesting to adults.

Originally Posted by rivviepop

Actually, that's interesting -- what happens if you force remove usb_storage module from the kernel (rmmod usb_storage) and *then* plug in the 7730? (maybe the module *thinks* it's mass storage by accident and is messing up?)

Got it! That wasn't it, but it pointed me in the right direction. First I downloaded me a copy of some live CD (Slax seemed nice) and verified the Blackberry would work there. Then I tried in Ubuntu's rescue mode - no luck. What I did notice though, while in rescue mode was this: berry_charge would be in the lsmod list whenever the BB was plugged in, no matter if I had removed it. That only stopped after killing udevd. When berry_charge was unloaded for good, the BB appeared in lsusb as well. Quickfix: mv berry_charge.ko BUGGERMENOLONGER-berry_charge.ko. Now it works like a charm and I could almost bang my head against a wall for hand-copying 252 contacts, 2 years worth of calendar entries and about 50 A4-pages of notes.

Thanks a lot rivviepop!

Only one question remains... is there a way to keep my new BB synchronized with Evolution and/or GoogleMail now?

(Oh, and I'll file a Ubuntu bug ASAP)

[rivviepop]

Originally Posted by rapha

Now it works like a charm and I could almost bang my head against a wall for hand-copying 252 contacts, 2 years worth of calendar entries and about 50 A4-pages of notes.

I guess it's too late to tell you about SyncML.

Only one question remains... is there a way to keep my new BB synchronized with Evolution and/or GoogleMail now?

I can't speak for Evolution, but GMail not so easily as they don't have a SyncML interface to your calendar or address book. Personally I use Mobical.net and the Nexthaus SyncML BB software to keep my device data on the website, there's a sticky here in the forum that tells you all about it.

[rapha]

Okay, just read that Sticky Thread... looks like an over-the-air solution, doesn't it. No worky for me then, as I don't have a data plan (won't happen anytime soon either). Looks like I might try to hack something together based on btool when time permits (next holidays maybe)... or has somebody else done such a script already? Since btool can download the handheld data it should only be a matter of converting it to something that Evolution or Thunderbird or whatever can understand. Converting the Notes to Tomboy would be the rockzors of cours

[rivviepop]

The folks on the barry-devel mailing list are very responsive and such; I'd recommend joining it and posting an email to see who's cooked up what; 0.9 apparently has a working opensync interface now so I expect more people are jumping in.