USB HDD and Flash Drives no longer recognized

Very odd indeed! I found a method to get the current kernel build to see the USB HDD box! The command line program lsusb seems to wake the kernel up to the fact that a USB HDD box is attached to the server. See the following:

userid@hostname:~$ lsusb
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
userid@hostname:~$ lsusb
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 002: ID 067b:3507 Prolific Technology, Inc. PL3507 ATAPI6 Bridge
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
userid@hostname:~$

The extra device in the second run is indeed the USB HDD box which the first lsusb some how managed to get something to wake up and actually see it.

Again these results are with the current build of the kernel: linux-image-2.6.32-24-generic-pae_2.6.32-24.43_i386.deb

The syslog entries found the found USB HDD box are as follows:

Sep 25 14:01:32 hostname kernel: [ 133.536034] usb 1-1: new high speed USB device using ehci_hcd and address 2
Sep 25 14:01:32 hostname kernel: [ 133.800427] usb 1-1: configuration #1 chosen from 1 choice
Sep 25 14:01:32 hostname kernel: [ 133.860147] Initializing USB Mass Storage driver...
Sep 25 14:01:32 hostname kernel: [ 133.860447] scsi5 : SCSI emulation for USB Mass Storage devices
Sep 25 14:01:32 hostname kernel: [ 133.860925] usbcore: registered new interface driver usb-storage
Sep 25 14:01:32 hostname kernel: [ 133.860936] USB Mass Storage support registered.
Sep 25 14:01:32 hostname kernel: [ 133.862617] usb-storage: device found at 2
Sep 25 14:01:32 hostname kernel: [ 133.862625] usb-storage: waiting for device to settle before scanning
Sep 25 14:01:37 hostname kernel: [ 138.860314] usb-storage: device scan complete
Sep 25 14:01:37 hostname kernel: [ 138.861189] scsi 5:0:0:0: Direct-Access Hitachi HDT725040VLAT80 V5CO PQ: 0 ANSI: 0
Sep 25 14:01:37 hostname kernel: [ 138.868478] sd 5:0:0:0: Attached scsi generic sg2 type 0
Sep 25 14:01:37 hostname kernel: [ 138.869259] sd 5:0:0:0: [sdb] Adjusting the sector count from its reported value: 781422768
Sep 25 14:01:37 hostname kernel: [ 138.869276] sd 5:0:0:0: [sdb] 781422767 512-byte logical blocks: (400 GB/372 GiB)
Sep 25 14:01:37 hostname kernel: [ 138.871773] sd 5:0:0:0: [sdb] Write Protect is off
Sep 25 14:01:37 hostname kernel: [ 138.871786] sd 5:0:0:0: [sdb] Mode Sense: 03 00 00 00
Sep 25 14:01:37 hostname kernel: [ 138.871794] sd 5:0:0:0: [sdb] Assuming drive cache: write through
Sep 25 14:01:37 hostname kernel: [ 138.873888] sd 5:0:0:0: [sdb] Adjusting the sector count from its reported value: 781422768
Sep 25 14:01:37 hostname kernel: [ 138.874884] sd 5:0:0:0: [sdb] Assuming drive cache: write through
Sep 25 14:01:37 hostname kernel: [ 138.875050] sdb: sdb1
Sep 25 14:01:37 hostname kernel: [ 138.893762] sd 5:0:0:0: [sdb] Adjusting the sector count from its repo...

Bug still exists / persists with the kernel update to linux-image-2.6.32-25-generic-pae_2.6.32-25.44_i386.deb this morning.

The lsusb command still works to get the server to recognize the attached / powered on USB HDD.

I am running 3 desktops where the USB mass storage (digital camera, USB sticks, etc.) is not recognized and does not show up in lsusb for the desktop I updated this morning to:

carl@cjfh3:~$ uname -a
Linux cjfh3 2.6.32-25-generic #44-Ubuntu SMP Fri Sep 17 20:26:08 UTC 2010 i686 GNU/Linux

The work-around did not work for me.

I tried booting on previous kernel, but that did not help.

Please disregard my entry #4
As it turns out a colleague told me that lightening had hit that particular computer and destroyed the internal USB. Thus, I cannot identify any bug after upgrade.
A very stupid note to Michael Lueck: There is something called usb-automount or something like that in the repositories. I take it that that might help you out until the issue is solved in a more proper way.

@Carl: That puts a whole new "light" on the issue you are seeing! ;-)

Yes I had come across that http://packages.ubuntu.com/lucid/usbmount package looking for possibilities as to why the drives were not being seen. I intend to use that package to try to mount drives to a consistent mount point no matter what device within /dev the drives show up as. On Ubuntu Server we have troubles that sometimes the drive decides to change devices from /dev/sdb to /dev/sdc.

However, the system still needs to see the USB drive in the first place! :-)

Same / Similar issue - USB was working (Flash drives and USB Canon i965 printer) & then stopped working...
Found turning off ehci allowed both to work....

cd /sys/bus/pci/drivers/ehci_hcd/
sudo sh -c 'find ./ -name "0000:00:*" -print| sed "s/\.\///">unbind'
(echos unbind to the ehci device id)

lots of 'unable to enumerate' messages in dmesg;

[ 1744.750662] usb 1-1: new high speed USB device using ehci_hcd and address 42
[ 1744.880420] usb 1-1: device descriptor read/64, error -71
[ 1745.120056] usb 1-1: device descriptor read/64, error -71
[ 1745.350173] usb 1-1: new high speed USB device using ehci_hcd and address 43
[ 1745.480055] usb 1-1: device descriptor read/64, error -71
[ 1745.723513] usb 1-1: device descriptor read/64, error -71
[ 1745.950062] usb 1-1: new high speed USB device using ehci_hcd and address 44
[ 1746.370083] usb 1-1: device not accepting address 44, error -71
[ 1746.490094] usb 1-1: new high speed USB device using ehci_hcd and address 45
[ 1746.910045] usb 1-1: device not accepting address 45, error -71
[ 1747.030065] usb 1-2: new high speed USB device using ehci_hcd and address 46
[ 1747.160070] usb 1-2: device descriptor read/64, error -71
[ 1747.407157] usb 1-2: device descriptor read/64, error -71
[ 1747.630059] usb 1-2: new high speed USB device using ehci_hcd and address 47
[ 1747.760053] usb 1-2: device descriptor read/64, error -71
[ 1748.000055] usb 1-2: device descriptor read/64, error -71
[ 1748.230102] usb 1-2: new high speed USB device using ehci_hcd and address 48
[ 1748.650063] usb 1-2: device not accepting address 48, error -71
[ 1748.770060] usb 1-2: new high speed USB device using ehci_hcd and address 49
[ 1749.190037] usb 1-2: device not accepting address 49, error -71
[ 1749.190074] hub 1-0:1.0: unable to enumerate USB device on port 2
[ 1749.370122] usb 1-3: new high speed USB device using ehci_hcd and address 50
[ 1749.500063] usb 1-3: device descriptor read/64, error -71
[ 1749.740065] usb 1-3: device descriptor read/64, error -71
[ 1749.973733] usb 1-3: new high speed USB device using ehci_hcd and address 51
[ 1750.100057] usb 1-3: device descriptor read/64, error -71
[ 1750.340061] usb 1-3: device descriptor read/64, error -71
[ 1750.570103] usb 1-3: new high speed USB device using ehci_hcd and address 52
[ 1750.990070] usb 1-3: device not accepting address 52, error -71
[ 1751.110118] usb 1-3: new high speed USB device using ehci_hcd and address 53
[ 1751.531122] usb 1-3: device not accepting address 53, error -71
[ 1751.531520] hub 1-0:1.0: unable to enumerate USB device on port 3
[ 1751.710204] usb 1-4: new high speed USB device using ehci_hcd and address 54
[ 1751.843296] usb 1-4: device descriptor read/64, error -71
[ 1752.090084] usb 1-4: device descriptor read/64, error -71
[ 1752.320105] usb 1-4: new high speed USB device using ehci_hcd and address 55
[ 1752.460086] usb 1-4: device descriptor read/64, error -71
[ 1752.690515] usb 1-4: device descriptor read/64, error -71
[ 1752.920088] usb 1-4: new high speed USB device using ehci_hcd and address 56
[ 1753.340041] usb 1-4: device not accepting address 56, error -71
[ 1753.460069] usb 1-4: new high speed USB device using ehci_hcd and a...

I tried the "unbind" script from David Mitchell does allow my system to "see" USB flash drives when I insert them. So does that point to the root cause?

BTW, I documented my system details and behavior in a similar bug report https://bugs.launchpad.net/ubuntu/+source/linux/+bug/669902.

Given that USB is a pretty fundamental function, and my searching indicates the problem is likely widespread, what has to happen now to get an owner assigned and the problem resolved?

@David Mitchell #7, Thank you for posting your work-around solution. That just helped me get a client's Western Digital 2.5" 500GB USB HDD to attach.

lsusb alone would not alert the computer to the drive's presence.
Your step allowed lsusb to see the drive.

I did notice that AFTER running your command, the "Linux Foundation 2.0 root hub" device disappears from the output of lsusb.

I checked my server again - indeed lsusb still shows the USB 2.0 device at the same time as the USB HDD.

The USB 2.0 controller disappearing on my client's server got me suspicious to an additional observation which I think might be related. Our backups are taking an EXTREMELY long time to complete. It is as if these USB 2.0 HDD's are actually connecting via USB 1.1 on Ubuntu 10.04 whereas on Ubuntu 9.10 and 9.04 there was no need to run additional commands and performance was snappy.

Backups usually took around an hour. The client's backup has been running for about 9 hours already and has only backed up 29GB of around 125GB.

It certainly smells like it is operating with a USB 1.1 connection to me.

Any clues as to why 10.04 is causing such difficulties? This is quite unacceptable... a real show stopper. I am probably going to have a screaming client in the morning when the backup is STILL running and they need to being the weeks tasks... probably scream to back level their server to Ubuntu 9.10, which is quite impossible.

Suggestions please?

I think there are two sub-issues going on with this trouble attaching USB mass storage devices to Ubuntu Server 10.04:

1) For our IOGear brand USB boxes, it is necessary to "wake up" Linux to the drive's presence attached to a USB port. I tested a backup last evening and it was speed-wise normal USB 2.0 expected performance. Looking at the lsusb output, indeed it is showing up on the USB 2.0 controller bus. This is a workable situation.

2) For our client who happens to backup to Western Digital "My Passport Essential" drives http://www.wdc.com/en/products/products.aspx?id=440 , the syntax specified by David Mitchell #7 does at least get the server to see the drive, however in that case it runs (crawls) at USB 1.1 performance which is not acceptable.

This really needs to get fixed in 10.04, and preferably SOON. What additional information could we provide to be of assistance?

aaahhh, could we get some assistance on this report please?

I do not have additional spare Intel Atom systems to test with, but would upgrading away from 10.04 LTS to 10.10 work where 10.04 is failing?

From the reports, seems several people are also using Intel Atom based boards.

Suggestions please? Our client is quite ticked with Ubuntu over this issue. Thank you.

Oh, my bad... it was in the Ubuntu Forums where someone responded with similar trouble using Intel Atom boards:

"10.04 server not recognizing USB drives suddenly"
http://ubuntuforums.org/showthread.php?t=1576834

Possibly fixed in lastest kernel revision - my box is now running

david@fitsserver:~$ uname -a
Linux fitsserver 2.6.32-26-server #48-Ubuntu SMP Wed Nov 24 10:28:32 UTC 2010 x86_64 GNU/Linux
david@fitsserver:~$ ^C
david@fitsserver:~$

I there are no usb errors (unable to enumerate) in dmesg

David

There appears to be someone else reporting this same sort of thing:

"external usb storage does not work with intel D945GCLF motherboard"
https://bugs.launchpad.net/bugs/625134

@David, seeing your mention of dmesg, I went checking my dmesg output and came upon that same error. Thus have commented on and subscribed to that other bug report.

Problem for us unchanged since #11.

@Michael Lueck: upgrading from Ubuntu 10.04 LTS to Ubuntu 10.10 worked for me. The USB drives can now be mounted. Have not tested the performance of them, but it seems acceptable.

#16 as per my post #14 is possibly (also) fixed by latest kernel update to 10.04 LTS...

David

@Davud #17

I am still getting nonsense errors with the current x86 10.04 kernel

[688424.636036] hub 1-0:1.0: Cannot enable port 1. Maybe the USB cable is bad?
[688427.604039] hub 1-0:1.0: Cannot enable port 1. Maybe the USB cable is bad?
[688430.572035] hub 1-0:1.0: Cannot enable port 1. Maybe the USB cable is bad?
[688433.540032] hub 1-0:1.0: Cannot enable port 1. Maybe the USB cable is bad?
[688433.540185] hub 1-0:1.0: unable to enumerate USB device on port 1
[688433.816034] usb 1-2: new high speed USB device using ehci_hcd and address 6
[688448.964027] usb 1-2: device descriptor read/64, error -110
[688464.216028] usb 1-2: device descriptor read/64, error -110
[688464.432030] usb 1-2: new high speed USB device using ehci_hcd and address 7
[688479.580526] usb 1-2: device descriptor read/64, error -110
[688494.832031] usb 1-2: device descriptor read/64, error -110
[688495.048040] usb 1-2: new high speed USB device using ehci_hcd and address 8

Did you happen to be using the x64 kernel?

Per #14, yes you are using x64... that might be the difference.

OK, "still a problem in the x86 kernel, Ubuntu please fix it soon!!! Seems you fixed it in the x64 kernel."

I just IPL'ed our server, issued sudo lsusb PRIOR to powering on the drive a couple of times, powered it on, and the USB drive attaches cleanly w/o errors that the cable might be bad or or or...

So on our server with IOGear USB HDD boxes, it is CRITICAL that lsusb be run PRIOR to attempting to attach... else you are just wasting your time. Again, per #19, "Ubuntu please fix it soon!!!"

Michael: I have been playing with this as well. Here is something for you to try as a workaround. I find that when I disable ehci, then re-enable it, I can see the USB key and the one high speed device I have (a Logitech C250 webcam) is found and driven with ehci driver. BTW, I'm using 10.04 desktop.

Here is the code sequence:

echo -n 0000:00:10.4 | sudo tee -a /sys/bus/pci/drivers/ehci_hcd/unbind

echo -n 0000:00:10.4 | sudo tee -a /sys/bus/pci/drivers/ehci_hcd/bind

(replace 0000:00:10.4 with whatever number is in /sys/bus/pci/drivers/ehci_hcd)

I grabbed this from http://ubuntuforums.org/showpost.php?p=8716597&postcount=4

This is still a bug, and a rather severe one in my opinion. It's not like this is some esoteric operation. I'm just trying to get a basic USB key to be recognized!

@John: Your steps appear to have gotten the drive connected via the USB 2.0 hub on my client's server. At least the output of lsusb shows it on the same "Bus 001" which is the bus of the USB 2.0 controller. I modified their "USB HDD fix script" to run these commands when they are mounting their backup drive. I will report again once we see if the backup runs at USB 2.0 performance... but so far, looking promising! Thank you! :-)

@John, thank you for following up with that suggestion. Client reports a snappy backup!

I worked up a bit of automation for the processes, as follows:

# fixext_backup.sh

#!/bin/bash -x

cd /sys/bus/pci/drivers/ehci_hcd

USBDRIVE=`/usr/bin/find ./ -name "0000:00:*"`
echo $USBDRIVE

echo $USBDRIVE | /bin/sed "s/\.\///">unbind
echo $USBDRIVE | /bin/sed "s/\.\///">bind

/usr/sbin/lsusb
/bin/sleep 5
/usr/sbin/lsusb

The first lsusb indeed does not find the drive, the second one after waiting five seconds does, and the client may carry on with their backup procedure at that point.

FYI, this is still an issue with the current 10.04 LTS x86 PAE kernel build.

Linux ldslnx01 2.6.32-40-generic-pae #87-Ubuntu SMP Mon Mar 5 21:44:34 UTC 2012 i686 GNU/Linux

I still must run my fixext_backup.sh script to get the USB HDD to attach via the USB 2.0 controller.

Any suggestions what is still broken in the 10.04 kernel?

Michael Lueck, thank you for reporting this and helping make Ubuntu better. Please be sure to confirm this issue exists with the latest development release of Ubuntu. ISO CD images are available from http://cdimage.ubuntu.com/daily/current/ . If the issue remains, please run the following command in the development release from a Terminal (Applications->Accessories->Terminal). It will automatically gather and attach updated debug information to this report.

apport-collect -p linux <replace-with-bug-number>

Also, if you could test the latest upstream kernel available that would be great. It will allow additional upstream developers to examine the issue. Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Once you've tested the upstream kernel, please remove the 'needs-upstream-testing' tag. This can be done by clicking on the yellow pencil icon next to the tag located at the bottom of the bug description and deleting the 'needs-upstream-testing' text.

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

If you are unable to test the mainline kernel, for example it will not boot, please add the tag: 'kernel-unable-to-test-upstream'.

Please let us know your results. Thanks in advance.

@Christopher,

I do not have capabilities to test the latest Ubuntu on our production servers. I will not be upgrading production servers until well after 12.04 LTS releases stable.

As indicated in this dupe report:
https://bugs.launchpad.net/bugs/625134

Upgrading to Ubuntu 10.10 resolved the problem for the OP, per post #3
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/625134/comments/3

So already it was corrected in the 10.10 kernel tree.

What ever the fix is it has never been identified and merged into the 10.04 LTS kernel tree.

Could you please put this bug report back to "Confirmed" against the 10.04 LTS version tree?

If there is a 10.04 LTS kernel build already prepared that I should test, please indicate so in a reply. I have tested other 10.04 kernel fixes already and the fixes have been distributed to production machines... so I am aware of how to do so. I recall different instructions in that case than "KernelMainlineBuilds", however.

Thank you!

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

@Ubuntu, I replied to your latest update, and along comes janitor and expires the bug report. Reopening hopefully. This is still an issue with the latest 10.04.4 x86 build.

A while ago I upgraded from Ubuntu Server 10.04 LTS to 12.04 LTS, and the USB HDD is properly attached automatically to a USB2 controller rather than USB 1.1. I no longer need the script posted in #23 to wake up the USB subsystem back to reality.

Michael Lueck, this bug report is being closed due to your last comment https://bugs.launchpad.net/ubuntu/+source/linux/+bug/645211/comments/29 regarding this being fixed with an update. For future reference you can manage the status of your own bugs by clicking on the current status in the yellow line and then choosing a new status in the revealed drop down box. You can learn more about bug statuses at https://wiki.ubuntu.com/Bugs/Status. Thank you again for taking the time to report this bug and helping to make Ubuntu better. Please submit any future bugs you may find.