On Wed, Sep 26, 2012 at 12:03:48AM +0300, Adrian Sandu wrote: > On Tue, Sep 25, 2012 at 9:26 AM, Adrian Sandu <dexter@xxxxxxxxxxx> wrote: > > On Tue, Sep 25, 2012 at 12:38 AM, Sarah Sharp > > <sarah.a.sharp@xxxxxxxxxxxxxxx> wrote: > I did some more tests. Just to make sure that I wasn't imagining things ... > > The file I'm trying to copy is a msdn iso of about 1.7G ( don't know > if this matters as I've tried with a dd if=/dev/urandom of=/root/en > bs=1024 count=2GB ) Ok, good to know. So you're using 1K block sizes. Were you copying the file with dd before, or just drag and dropping it? > just a reminder .. my laptop is a vaio vpcf13m1e ( which seems to work > every time ) .. the one with the problems is an asrock 152d .. Both > seem to share the usb3 root hub chipset ( a nec uPD720200 ) > > the tests are done using a gembird usb3 hub ( which reports > idVendor=2109, idProduct=0810 VIA Labs, Inc. 4-Port USB 3.0 HUB .. > idVendor=2109, idProduct=3431 .. USB2.0 Hub ) > I have another hub, a manhattan ( which reports the same 2 stuff at lsusb ) > > sda3 is my local drive in the asrock ( a 2.5" normal hdd which does > its job just fine for > 1 year now 24/7 ) > > 1) Copy to/from the drive directly from/to my laptop 2 times ( in each > dirrection ) > WORKS > 2) Copy to/from and from/to 1 drive in/through the hub on my laptop 2 > times ( in each dirrection ). > WORKS > 3) Copy to/from the drive directly from/to the asrock 2 times ( in > each dirrection ) > WORKS > 4) Copy to/from and from/to 1 drive in/through the hub on my asrock 2 > times ( in each dirrection ). > > I have /dev/sdb1 mounted in /media/sdg1 .. (sdg because I usually have > more drives) > cp /media/sdg1/en2 ~/en2 > try1 copy from sdb1 to sda3 .. seems to succeed ?! cp didn't say > anything .. but /var/log/messages shows some stuff > try2 copy from sdb1 to sda3 .. seems to succeed ?! cp didn't say > anything .. just as above > > cp ~/en2 /media/sdg1/en2 > try3 copy from sda3 to sdb1 .. seems to succeed .. cp didn't say > anything .. nor anything weird in /var/log/messages > try4 .. just plugging in the drive caused some messages .. didn't have > the cat on 9u started though and couldn't reproduce it .. > try5 copy from sda3 to sdb1 .. seems to succeed .. cp didn't say > anything .. nor anything weird in /var/log/messages > > copy using mc from /media/sdg1/en2 to ~/en2 > try6 .. copy from sdb1 to sda3 .. seems to succeed .. but I see > messages in /var/log/messages .. > try7 .. same .. Ok, so reads from the drive produce some messages, but writes don't. Note that "weird stuff in /var/log/messages" does not necessarily mean that the hard drive is broken. Sometimes it may take two tries to make a particular read or write succeed, or maybe the drive doesn't like a particular SCSI command type and so the SCSI layer tries another command type. The device sometimes gets reset, but that's normal. And if the messages you mention are about disabling endpoints or LPM, then those are just warnings, not errors. Basically, your file transfer should be OK as long as you don't see any I/O errors. What really matters is whether the file got successfully copied or not. If cp didn't give you any errors, the file got copied. You can make sure it's exact duplicate by running diff against the two files. Or if you're copying a large number of directories, you could use git in the base directory, create the initial commit of the files, copy the directory over, and run `git diff` to see if anything changed. > copy using mc from ~/en2 to /media/sdg1/en2 > try8 .. copy from sda3 to sdb1 .. succedes just fine .. no messages no > nothing .. > try9 .. same .. > > adding a new drive in the hub ! :) > copy using mc from sdb to sdc > try10 .. there's some weird thing going in my /var/log/messages .. > > plugging out and back in the hub with all drives :) ( along with some > rmmod xhci_hcd .. usb_storage and modprobe-ing them again ) fails .. Is there any way you could send the logs as attachments or get your mail client not to wrap long lines? It's pretty hard to read. > try11 .. 2012-09-26T00:02:38.810382+03:00 d3xt3r01 kernel: > [93320.454212] xhci_hcd 0000:04:00.0: Host not halted after 16000 > microseconds. > 2012-09-26T00:02:38.810393+03:00 d3xt3r01 kernel: [93320.454212] > xhci_hcd 0000:04:00.0: Host controller not halted, aborting reset. Hmm, that's not cool. The xHCI hardware didn't want to halt within 16 seconds, which usually means it's hung. I'm guessing the modprobe or rmmod failed for the xhci-hcd driver? > 2012-09-26T00:02:38.815117+03:00 d3xt3r01 kernel: [93320.496220] > xhci_hcd 0000:04:00.0: USB bus 1 deregistered > 2012-09-26T00:02:41.416854+03:00 d3xt3r01 kernel: [93323.097171] > xhci_hcd 0000:04:00.0: xHCI Host Controller > 2012-09-26T00:02:41.416932+03:00 d3xt3r01 kernel: [93323.097927] > xhci_hcd 0000:04:00.0: new USB bus registered, assigned bus number 1 > 2012-09-26T00:02:41.463589+03:00 d3xt3r01 kernel: [93323.143244] > xhci_hcd 0000:04:00.0: Host not halted after 16000 microseconds. > 2012-09-26T00:02:41.463629+03:00 d3xt3r01 kernel: [93323.143366] > xhci_hcd 0000:04:00.0: can't setup > 2012-09-26T00:02:41.463636+03:00 d3xt3r01 kernel: [93323.143381] > xhci_hcd 0000:04:00.0: USB bus 1 deregistered > 2012-09-26T00:02:41.463647+03:00 d3xt3r01 kernel: [93323.143836] > xhci_hcd 0000:04:00.0: init 0000:04:00.0 fail, -110 > 2012-09-26T00:02:41.463698+03:00 d3xt3r01 kernel: [93323.143852] > xhci_hcd: probe of 0000:04:00.0 failed with error -110 > 2012-09-26T00:02:48.399042+03:00 d3xt3r01 kernel: [93330.080709] usb > 3-2: USB disconnect, device number 13 > 2012-09-26T00:02:50.344948+03:00 d3xt3r01 kernel: [93332.026075] usb > 3-1: new high-speed USB device number 18 using ehci_hcd > 2012-09-26T00:02:50.460032+03:00 d3xt3r01 kernel: [93332.141141] usb > 3-1: New USB device found, idVendor=2109, idProduct=3431 > 2012-09-26T00:02:50.460041+03:00 d3xt3r01 kernel: [93332.141146] usb > 3-1: New USB device strings: Mfr=0, Product=1, SerialNumber=0 > 2012-09-26T00:02:50.460067+03:00 d3xt3r01 kernel: [93332.141152] usb > 3-1: Product: USB2.0 Hub > 2012-09-26T00:02:50.460095+03:00 d3xt3r01 kernel: [93332.141646] hub > 3-1:1.0: USB hub found > 2012-09-26T00:02:50.460100+03:00 d3xt3r01 kernel: [93332.141895] hub > 3-1:1.0: 4 ports detected > 2012-09-26T00:02:50.732921+03:00 d3xt3r01 kernel: [93332.414276] usb > 3-1.2: new high-speed USB device number 19 using ehci_hcd > 2012-09-26T00:02:51.204932+03:00 d3xt3r01 kernel: [93332.886035] usb > 3-1.2: device not accepting address 19, error -71 > 2012-09-26T00:02:51.215940+03:00 d3xt3r01 kernel: [93332.897641] hub > 3-1:1.0: unable to enumerate USB device on port 2 > 2012-09-26T00:02:51.227932+03:00 d3xt3r01 kernel: [93332.909636] hub > 3-1:1.0: unable to enumerate USB device on port 3 > 2012-09-26T00:02:51.291054+03:00 d3xt3r01 kernel: [93332.972259] usb > 3-1.4: new high-speed USB device number 22 using ehci_hcd > 2012-09-26T00:02:51.313950+03:00 d3xt3r01 kernel: [93332.995654] hub > 3-1:1.0: unable to enumerate USB device on port 4 > > now I need to reboot .. > > After each test I've only plugged ou the drive from the hub ( the hub > remained connected to the computer ) > Results: Directory indexing enabled .. http://d3xt3r01.tk/~dexter/usbmon/ > > The kernel used was vanilla 3.4.11 on a gentoo system .. > After all these tests .. I think the drive is ok .. at least .. the > hub might have some problems ( both the gembird and the one in the > asrock .. or at least one of them .. ) but I don't get what as .. it > seems to work on my laptop :| > > Please let me know how I can be of more assistance / help :| What I actually need to know is what "weird stuff" appeared in the logs during the disk reads. I suspect your usbmon log is going to be pretty big, and I'd like to get a smaller sample of any real disk error. Sarah Sharp -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
