So I’ve got a weird issue with my usb external drive. It’s a generic enclosure in to which I’ve mounted a toshiba IDE drive. It appears as “Generic USB to IDE bridge”.
Said drive works fine on every machine I try to use it on, but NOT my laptop.
Oh, it will appear in LSUSB, and in dmesg, (sometimes it appears, but the partitions do not, e.g. I see sdc, but not sdc1 and sdc2).
However, on any attempt to access it (it mounts fine too), Dolphin hangs (and so does most of KDE). I assume a disk sleep is the problem. Unplugging the drive resolves the hang.
Anyway, ages later, the drive will copy bits of data in small spurts, until it hangs for some more time. After a while, I see an ACHI/USB reset message in the system logs, and the drive resumes copying for some more time.
The drive is bus powered, but it doesn’t matter if I use one/both USB connectors, and like I said, it works fine on my desktop, which also has SuSE 11.4 on it.
Any ideas what’s going on? Sorry if I left out some info, I wanted to get this out and let you folks stew over it while I’m running some errands.
i’d suspect that laptop has some sort of hardware failure…or perhaps
in need of a bios update (check the makers web site)…
i mean the USB drive works everywhere else, so the problem must be
unique to the laptop hardware, right?
unless of course you have some reason to suspect the software/settings
of the 11.4 on that machine has been damaged, somehow…do you? anything
else ‘weird’?
so, how about trying a different usb enclosure/disk with that
laptop…if a different toshiba IDE drive works ok, then . . .
> Sorry if I left out some info, I wanted to
> get this out and let you folks stew over it while I’m running some
> errands.
how about these details, most (but maybe not all) could impact the
answer and/or help focus in on the right path to the answer:
I’ll give the BIOS update a try. Other USB drives/devices work fine, so it’s not a port failure. I’ve always figured the lappy and drive just don’t play nice together. If the BIOS doesn’t help, I do have another IDE drive I can stick in the enclosure and see what happens. I’ll post back with additional details or to let you know if it is resolved.
So I’ve got a weird issue with my usb external drive. It’s a generic enclosure in to which I’ve mounted a toshiba IDE drive. It appears as “Generic USB to IDE bridge”.
Said drive works fine on every machine I try to use it on, but NOT my laptop.
Oh, it will appear in LSUSB, and in dmesg, (sometimes it appears, but the partitions do not, e.g. I see sdc, but not sdc1 and sdc2).
Strange behaviour can occur when there is insufficient power from the USB port available for the external device. Perhaps that is the problem you’re seeing here.
BIOS update does not seem to have fixed the issue. I should also note that if I try the drive in coughXPcough on the same machine, I get the infamous “this device cannot start”.
No luck changing the Legacy USB setting, and my BIOS has little else to fiddle with. No other USB settings.
FYI, the USB detection in dmesg looks normal, except I get a lot of “usb 1-1: Reset high speed usb device using ehci_hcd and address 5”
Don’t know the exact specs for the system off the top of my head, but it is an MSI M662 laptop (AKA MS1034)
Running SuSE 11.4 32-bit. This problem has been going on for quite a while now, I just kinda ignored it. (i.e. it’s been there since previous releases, and I can’t recall if it ever worked properly).
So it’s definitely a hardware issue.
I don’t have my other IDE laptop drive on hand, so I can’t test much else at this point, unless there are other suggestions here. USBID of the device is 058f:6391, perhaps someone’s google-fu is better than mine and can find a kernel bug report on it or so.
BIOS update does not seem to have fixed the issue. I should also note that if I try the drive in coughXPcough on the same machine, I get the infamous “this device cannot start”.
This would be consistent with a lack of power from the USB. If you must, try with a LiveCD (any distro, version) to confirm…
A possible workaround would be to use a powered USB hub to see if that makes a difference.
Thought that myself many times, but no success, I’m afraid. The enclosure came with a Y-cable so it can draw power from two ports if necessary, and it makes no difference if I connect one or both cables. (be they to a hub that I know can provide 500mA per port or my laptop).
It will actually work on a single port on any of the other machines I’ve used it on.
(i didn’t read every word again, but) i think you said it is a drive in
an enclosure…so, if it can be, open the enclosure and unplug the
drive…if it is possible to replace the usb cable between the computer
and the enclosure, do that…then put it all back together and see if
magic has happened (the thought being that maybe it is just a bit of
corrosion in the cable contact spots…or in the cable or its plugs
itself…
oh, just in case you brick it, please read my caveat before you take it
apart.
–
DD Caveat
openSUSE®, the “German Automobiles” of operating systems
Already done that, no luck I’m afraid. I’ll get my other laptop drive and test it in the enclosure as well, to see if it’s an issue with the drive or the adapter. That will take a few days though, I’m afraid.
I’m sorry, what? This was not something I’d done as of the initial few posts, it came about during the other troubleshooting, where I took the drive out of the enclosure for a visual inspection and to see if it was making any unusual noises. Thus, it was not an “already done that” as in “oh, I did that ages ago”, it was as in “oh, that’s something I’ve just tried and didn’t immediately post because this forum is not my blog/life storyboard”.
Ok, so I had another go in XP trying to get the drive to work. Whilst derping around in the device manager (didn’t give a Cannot start, but the partitions didn’t appear) I noticed my Wifi card connects via usb (despite occupying a mini-PCI slot). On a hunch, I pushed the wifi kill button, and BOOM, my partitions appeared and I was able to copy a 600MB ISO without issue. As soon as I enable the wifi again, it grinds to a halt.
Linux performance is also much better with the wifi off, but it still halts occasionally (likely differences in how the wifi-off is handled).
The card (Ralink RT73) has a push-button kill switch, and doesn’t actually dissappear from lsusb, so it’s not as though it’s cutting the power to the card.
I also don’t think it is a USB bandwidth issue, since there’s very little data actually being transmitted when the connection is in use.
Is it possible for a usb device to just hog the bus like that?
(I’ll try seeing what happens if I kill the wifi but enable the USB webcam, and post back with those results.)
Update: As soon as I connect any other usb device (webcam, flash drive, usb mouse) the external drive will start hiccuping. This gets weirder and weirder, since it works fine on my desktop (which always has a usb mouse and printer connected) and other usb storage devices will work fine on my laptop, even if the wireless is on.
Looks like we have a chipset and hardware specific firmware or BIOS bug on our hands?
On 09/16/2011 06:56 PM, VintagePC wrote:
>
> Looks like we have a chipset and hardware specific firmware or BIOS bug
> on our hands?
or, i’d put a multi-meter on those USB ports because (as Deano said)
“Strange behaviour can occur when there is insufficient power from the
USB port available for the external device.”
–
DD
openSUSE®, the “German Automobiles” of operating systems
I can certainly try that, but I don’t have access to my DMM at the moment.
Would connecting through a wall-powered USB hub offer a similar test, (since the hub doesn’t draw power from the laptop), or do you think this is a case where the other device is simply dragging down the other ports to a point where it becomes unusable for the HDD?
On another note, my laptop suddenly sprouted (well after this USB issue) a problem where the num-lock LED is really dim… perhaps I’m just witnessing it dying a slow and painful death
On 09/16/2011 09:06 PM, VintagePC wrote:
>
> I can certainly try that, but I don’t have access to my DMM at the
> moment.
actually (now that i think about it), the meter might read volts spot on
but not tell you anything about the amps available…so, if it is
delivering the right volts but enough enough available current draw, the
meter tells nothing useful…
> Would connecting through a wall-powered USB hub offer a similar test,
> (since the hub doesn’t draw power from the laptop)
i think so…
> On another note, my laptop suddenly sprouted (well after this USB
> issue) a problem where the num-lock LED is really dim… perhaps I’m
> just witnessing it dying a slow and painful death
this could absolutely be symptoms of a dying motherboard…
a cracked board, some loose or corroded connections…etc etc…
–
DD Caveat
openSUSE®, the “German Automobiles” of operating systems
> actually (now that i think about it), the meter might read volts spot on but not
> tell you anything about the amps available…so, if it is delivering the right
> volts but enough enough available current draw, the meter tells nothing useful…
In every situation I have ever seen, a power supply that is unable to supply the
necessary current will also have a voltage drop. The voltage measurement should
be useful.
