[Q] NC / Win7-64 / ADB

---

Since there seems to be a number of people fighting to get ADB working with the NC on Win7-64 machines, I thought I'd start a dedicated thread to get to the bottom of this issue.

In my case, ADB is working from my laptop but not my desktop. At various stages I have different devices, including an "Android Composite ADB Interface" and and "Android ADB Interface". At no point does `adb devices` return anything.

Here's some bits that will hopefully be helpful in troubleshooting.

Without the drivers installed (more accurately, with them uninstalled and deleted), I have the following devices in Device Manager:

Other devices -> Nook
Hardware Ids:
USB\VID_2080&PID_0002&REV_0216&MI_01
USB\VID_2080&PID_0002&MI_01
Universal Serial Bus controllers -> USB Composite Device
Hardware Ids:
USB\VID_2080&PID_0002&REV_0216
USB\VID_2080&PID_0002
Universal Serial Bus controllers -> USB Mass Storage Device
Hardware Ids:
USB\VID_2080&PID_0002&REV_0216&MI_00
USB\VID_2080&PID_0002&MI_00

Could someone with a working ADB under Win7-64 uninstall their drivers and post the above information? If it's the same, could you zip your usb_drivers dir and upload that?

ADB version from SDKr08:

C:\>adb version
Android Debug Bridge version 1.0.26

ADB logcat with debug all:

C:\>set ADB_TRACE=all

C:\>adb -d logcat
_adb_connect: host:version
socket_loopback_client: could not connect to tcp:5037
* daemon not running. starting it now on port 5037 *
* daemon started successfully *
_adb_connect: shell:export ANDROID_LOG_TAGS="" ; exec logcat
socket_loopback_client: port 5037 type tcp => fd 101
writex: 101 0x28dcd2 4: 30303132  0012
writex: 101 ok
writex: 101 0x28dc92 18: 686f73743a7472616e73706f72742d75  host:transport-u
writex: 101 ok
Switch transport in progress
readx: 101 0x28dc57 4
readx: 101 ok: 4641494c  FAIL
readx: 101 0x28dc57 4
readx: 101 ok: 30303130  0010
readx: 101 0x4221a0 16
readx: 101 ok: 646576696365206e6f7420666f756e64  device not found
adb_close: 101(lo-client:5037)
Switch transport failed
- waiting for device -
_adb_connect: host:version
socket_loopback_client: port 5037 type tcp => fd 102
writex: 102 0x28baf7 4: 30303063  000c
writex: 102 ok
writex: 102 0x41d3ed 12: 686f73743a76657273696f6e  host:version
writex: 102 ok
readx: 102 0x28ba77 4
readx: 102 ok: 4f4b4159  OKAY
readx: 102 0x28bb34 4
readx: 102 ok: 30303034  0004
readx: 102 0x28bb34 4
readx: 102 ok: 30303161  001a
adb_close: 102(lo-client:5037)
_adb_connect: host-usb:wait-for-usb
socket_loopback_client: port 5037 type tcp => fd 103
writex: 103 0x28baf7 4: 30303135  0015
writex: 103 ok
writex: 103 0x28cc01 21: 686f73742d7573623a776169742d666f  host-usb:wait-fo
writex: 103 ok
readx: 103 0x28ba77 4
readx: 103 ok: 4f4b4159  OKAY
readx: 103 0x28bb97 4

Again, if we can get a working environment's output, that would be great.

Those who are having the same problem, post the results of the above command here, too. Don't forget the `set ADB_TRACE=all`. We can see if there's any differences in the behavior in the non-working envs.

I should note that some people have been successful by removing and deleting the drivers, un/re-plugging the NC and letting Windows auto-install the drivers (if I'm understanding the process).

Originally Posted by edison

Since there seems to be a number of people fighting to get ADB working with the NC on Win7-64 machines, I thought I'd start a dedicated thread to get to the bottom of this issue.

In my case, ADB is working from my laptop but not my desktop. At various stages I have different devices, including an "Android Composite ADB Interface" and and "Android ADB Interface". At no point does `adb devices` return anything.

Here's some bits that will hopefully be helpful in troubleshooting.

Without the drivers installed (more accurately, with them uninstalled and deleted), I have the following devices in Device Manager:

Other devices -> Nook
Hardware Ids:
USB\VID_2080&PID_0002&REV_0216&MI_01
USB\VID_2080&PID_0002&MI_01
Universal Serial Bus controllers -> USB Composite Device
Hardware Ids:
USB\VID_2080&PID_0002&REV_0216
USB\VID_2080&PID_0002
Universal Serial Bus controllers -> USB Mass Storage Device
Hardware Ids:
USB\VID_2080&PID_0002&REV_0216&MI_00
USB\VID_2080&PID_0002&MI_00

Could someone with a working ADB under Win7-64 uninstall their drivers and post the above information? If it's the same, could you zip your usb_drivers dir and upload that?

ADB version from SDKr08:

C:\>adb version
Android Debug Bridge version 1.0.26

ADB logcat with debug all:

C:\>set ADB_TRACE=all

C:\>adb -d logcat
_adb_connect: host:version
socket_loopback_client: could not connect to tcp:5037
* daemon not running. starting it now on port 5037 *
* daemon started successfully *
_adb_connect: shell:export ANDROID_LOG_TAGS="" ; exec logcat
socket_loopback_client: port 5037 type tcp => fd 101
writex: 101 0x28dcd2 4: 30303132  0012
writex: 101 ok
writex: 101 0x28dc92 18: 686f73743a7472616e73706f72742d75  host:transport-u
writex: 101 ok
Switch transport in progress
readx: 101 0x28dc57 4
readx: 101 ok: 4641494c  FAIL
readx: 101 0x28dc57 4
readx: 101 ok: 30303130  0010
readx: 101 0x4221a0 16
readx: 101 ok: 646576696365206e6f7420666f756e64  device not found
adb_close: 101(lo-client:5037)
Switch transport failed
- waiting for device -
_adb_connect: host:version
socket_loopback_client: port 5037 type tcp => fd 102
writex: 102 0x28baf7 4: 30303063  000c
writex: 102 ok
writex: 102 0x41d3ed 12: 686f73743a76657273696f6e  host:version
writex: 102 ok
readx: 102 0x28ba77 4
readx: 102 ok: 4f4b4159  OKAY
readx: 102 0x28bb34 4
readx: 102 ok: 30303034  0004
readx: 102 0x28bb34 4
readx: 102 ok: 30303161  001a
adb_close: 102(lo-client:5037)
_adb_connect: host-usb:wait-for-usb
socket_loopback_client: port 5037 type tcp => fd 103
writex: 103 0x28baf7 4: 30303135  0015
writex: 103 ok
writex: 103 0x28cc01 21: 686f73742d7573623a776169742d666f  host-usb:wait-fo
writex: 103 ok
readx: 103 0x28ba77 4
readx: 103 ok: 4f4b4159  OKAY
readx: 103 0x28bb97 4

Again, if we can get a working environment's output, that would be great.

Those who are having the same problem, post the results of the above command here, too. Don't forget the `set ADB_TRACE=all`. We can see if there's any differences in the behavior in the non-working envs.

I should note that some people have been successful by removing and deleting the drivers, un/re-plugging the NC and letting Windows auto-install the drivers (if I'm understanding the process).

Attached is my usb_drivers folder from my working Windows 7 Pro x64 installation.

Note that this is the same usb driver zip from the nookcolor portal (http://nookdevs.com/NookColor_USB_ADB) under the 'windows' section, just with the necessary B&N product lines already added.

It's working!

Unfortunately, I'm not 100% sure which step I did make it work, but I list everything I can recall.

I'll first answer a few of my own questions in the hopes it'll help others.

Output of `adb -d logcat`:

C:\>adb -d logcat > logcat.txt
_adb_connect: host:version
socket_loopback_client: could not connect to tcp:5037
_adb_connect: shell:export ANDROID_LOG_TAGS="" ; exec logcat
socket_loopback_client: port 5037 type tcp => fd 101
writex: 101 0x28dcd2 4: 30303132  0012
writex: 101 ok
writex: 101 0x28dc92 18: 686f73743a7472616e73706f72742d75  host:transport-u
writex: 101 ok
Switch transport in progress
readx: 101 0x28dc57 4
readx: 101 ok: 4f4b4159  OKAY
Switch transport success
writex: 101 0x28dcd7 4: 30303265  002e
writex: 101 ok
writex: 101 0x28dde1 46: 7368656c6c3a6578706f727420414e44  shell:export AND
writex: 101 ok
readx: 101 0x28dc57 4
readx: 101 ok: 4f4b4159  OKAY

I had to redirect to the txt file because the screen was filled with a bunch of the following lines:

I/ActivityManager(  822): Starting activity: Intent { flg=0x10000000 cmp=android/com.android.internal.app.UMSServerActivity }

D/DeviceManagerBroadcastReceiver( 1011): action (android.intent.action.BATTERY_CHANGED)

D/WifiService(  822): ACTION_BATTERY_CHANGED pluggedType: 0

I've attached the usb_drivers I used, although I can't see any difference in them and the ones I had before.

Now for the stuff I did in-between my first post and this one (from memory).

1. I plugged the USB cable directly into the ports on my motherboard, instead of my PCI-X internal USB2.0 card. This was in response to suggestions on the other Win7 thread that you remove USB hubs from the mix. I'll try again with it plugged into the card later and report back.
   
   I should note that in order to do this, I had to unplug my HP Deskjet which has a card reader slot. I had two USB Mass Storage and USB Composite devices listed in Device Manager prior to removing this - one each for the NC and the HP. After the removal, only a single set. Dunno if it made any difference, but it's worth mentioning. I'll try it with that plugged in, too, and report back.
   
   
2. I uninstalled the 'Android Composite ADB Interface' from under 'Android phone' from Device Manager, without checking the "Delete driver software" box. Obviously at this point I had the drivers installed. To be honest, I can't recall which of the native (Nook, USB Composite Device, USB Mass Storage Device) I updated the drivers on, but I *believe* it was the "Nook".
   
   
3. After the above step, I unplugged the USB cable and plugged it back in. This was based on several suggestions to that effect. I repeated these two steps multiple times but nothing ever showed up in `adb devices`, so I tried rebooting.
   
   
4. During this and all subsequent reboots, I hit F8 on startup and selected the "Disable Driver Signature Enforcement".
   
   
5. Still nothing after the reboot, so I uninstalled the 'USB Composite Devices'. Windows informed me a reboot was required. Whoah! That was new!
   
   
6. After the 2nd reboot, Windows automatically installed drivers (presumably for the device I had just removed), but again required a reboot.
   
   
7. Since the required rebooting behavior was so different than previous attempts (none of which asked to boot), I tried `adb devices` again.
   
   Nothing.
   
   Drat!
   
   
8. Then I checked the c:\Users\edison\.android\adb_usb.ini file that I created yesterday. All it had was this!

   # ANDROID 3RD PARTY USB VENDOR ID LIST -- DO NOT EDIT.
   # USE 'android update adb' TO GENERATE.
   # 1 USB VENDOR ID PER LINE.

   I *know* I had inserted a single '0x2080' in the file with no spaces or CRs. I had double-, triple- and quadruple-checked that stinkin' file during the last two days of frustration and it always existed and contained the right string. Something had to have changed it, but I'm not sure what it was. At one point, I had run the SDK Manager (from \androidsdk) and removed the non-2.1-related packages, including the "USB drivers package, revision 4". That's my best guess for the culprit.
   
   So I added the "0x2080" to the adb_usb.ini file (after the 3 existing commented lines) and `adb-remount` gave me the sought-after "remount failed: No such file or directory", after which `adb devices` listed my NC!
   
   Woo.
   
   Hoo.
   
   FWIW, I don't know that `adb remount`, specifically, is necessary. I ran `adb kill-server`, then `adb start-server` afterwards and it worked that way as well. I think the key is restarting the server once everything else is in place.

Since there are SO many others having this exact problem, can those of you who are and who try these steps, record what you do and report back?

My guess is that rebooting without the signed drivers req, removing the 'USB Composite Device' and the subsequent reboots/auto driver installs is what worked.

I do want to clarify that I know my NC was properly rooted as I was able to sideload apks from my laptop. This was only an ADB issue on my PC. IMO, the best way to confirm your NC is rooted is to boot with the nooter SD card installed. If you get a standard nook boot process, it's not rooted. If the device shows up in Windows (as *anything*), but the screen stays blank, you're rooted. I was pretty careful on my SD boots, as per the guides, to wait long enough for the root to "take" (2 full mins).

I sincerely hope this helps as I've spent all of last night and today working on this and I hope to save as many others from the same rage-inducing frustrations as possible.

Thanks to everyone who has put in the effort to make this possible, and everyone who has stuck by the community and taken the time to help the rest of us out!

Originally Posted by edison

It's working!

Unfortunately, I'm not 100% sure which step I did make it work

....
[*]Then I checked the c:\Users\edison\.android\adb_usb.ini file that I created yesterday. All it had was this!

# ANDROID 3RD PARTY USB VENDOR ID LIST -- DO NOT EDIT.
# USE 'android update adb' TO GENERATE.
# 1 USB VENDOR ID PER LINE.

I *know* I had inserted a single '0x2080' in the file with no spaces or CRs. I had double-, triple- and quadruple-checked that stinkin' file during the last two days of frustration and it always existed and contained the right string. Something had to have changed it, but I'm not sure what it was. At one point, I had run the SDK Manager (from \androidsdk) and removed the non-2.1-related packages, including the "USB drivers package, revision 4". That's my best guess for the culprit.

So I added the "0x2080" to the adb_usb.ini file (after the 3 existing commented lines) and `adb-remount` gave me the sought-after "remount failed: No such file or directory", after which `adb devices` listed my NC!

Woo.

Hoo.
[/B]

this is what fixed it, it is a known issue that when the sdk updates using the sdk manager it replaces the adb_usb.ini with this file. you can manually edit it or just run 'android update adb' in command prompt to fix it. just fyi

I've never had an issue with adb_usb.ini being altered. I always check it before another adb attempt.

I went through the lists posted in here and came up with the same result on my win 7 64 bit setup. Disabling driver signing, uninstalling usb composite as well as removing the android composite adb interface and reinstalling.

I still need a way to confirm my nooter img has written correctly. There is no way to confirm this other than following the steps very precisely.

It just gets me angry *kicks nook*.

I had the ini file get changed after I uninstalled the drivers and reinstalled them.

Originally Posted by madsci73915

I had the ini file get changed after I uninstalled the drivers and reinstalled them.

Un-/re-installed the drivers via Device Mangler or the SDK Manager?

Originally Posted by moovius

I've never had an issue with adb_usb.ini being altered. I always check it before another adb attempt.

Same here. That's why I was so mystified when it changed this time.

Originally Posted by moovius

I went through the lists posted in here and came up with the same result on my win 7 64 bit setup. Disabling driver signing, uninstalling usb composite as well as removing the android composite adb interface and reinstalling.

Remember, I was going from memory, which was pretty fried after the previous 12 hours of fiddling. There may be another combination or step I missed.

Originally Posted by moovius

I still need a way to confirm my nooter img has written correctly. There is no way to confirm this other than following the steps very precisely.

Does your NC wake up to a blank screen when booting from the SD card?

To check the actual SD card, pop it back in your PC. It should have a regular 40MB FAT partition that Win7 can read. If you want, you can use DISKPART from the shell to verify the partition. It should go something like this:

C:\> DISKPART
DISKPART> list disk
DISKPART> select disk N <-- your SD card from the list above
DISKPART> list partition

You should see the 40MB partition listed, although I don't have a screen grab to show. I just reformatted my SD card 'cause the Kindle app requires one and I didn't have an extra spare, but I can either pick one up today or re-image it

Originally Posted by moovius

It just gets me angry *kicks nook*.

Hang in there, moovius, we'll figure this out! I *completely* understand your frustration, but persistence and tenacity will pay off!

OOC, do you have another computer you can try this on?

Originally Posted by edison

Hang in there, moovius, we'll figure this out! I *completely* understand your frustration, but persistence and tenacity will pay off!

OOC, do you have another computer you can try this on?

I had one of the guys from #nookdevs remoting my computer and even he was stumped.

I gave up on windows and downloaded Ubuntu Livecd:

1. From Ubuntu, I re-did the nooter img on the nook to make sure that it wasnt the issue behind all of this.

2. I spent 2-3 hours being instructed on how to work Ubuntu and run commands from the terminal. (That wiki really assumes a lot, it's hardly a step-by-step method, even this Linux expert was stumped with some of the instructions.)

3. Finally was able to see the nook with the adb device listing command.

4. Loaded z4 for Super User privileges, and added 5-6 more apps to test out.

5. Restarted my computer while playing around with the new programs on the nook.

6. Tried accessing the nook through win7 again, but no luck. Uninstalled the driver, etc, tried several things, still wouldnt work from windows. Tried from my win7 netbook also, but no dice.

Something either in my win7 installs or in my computer configurations is preventing communication between pc and nook color.

At least Ubuntu works.

pdanet

---

the drivers from pdanet seem to work just fine and will automatically install, never has used pdanet application though.
bg
http://www.junefabrics.com/android/
just download, don't buy, and install