On 1/23/17 10:12 , Robert Mustacchi wrote: > On 1/19/17 8:27 , Jason Lawrence wrote: >> On Fri, Jan 6, 2017, at 04:55 PM, Robert Mustacchi wrote: >>> On 1/5/17 18:43 , Jason Lawrence wrote: >>>> On Mon, Jan 2, 2017, at 06:14 PM, Robert Mustacchi wrote: >>>>> On 1/2/17 14:16 , Jason Lawrence wrote: >>>>>> I've upgraded to the 20161222T003450Z build and have been having issues >>>>>> with getting USB 3.0 drives recognized. The drives work fine on USB 2.0 >>>>>> ports. >>>>> >>>>> Hi Jason, >>>>> >>>>> Thanks for the report. Sorry you're seeing some trouble with this. The >>>>> key from the log is the bit: >>>>> >>>>> "WARNING: /pci@0,0/pci15d9,811@14 (xhci0): Connecting device on port 12 >>>>> failed" >>>>> >>>>> This indicates that something in the connection process failed. While >>>>> this is a bit of a complicated state machine, I have a D script that >>>>> should help us figure out what's going on. Here's the D script. >>>>> >>>>> https://us-east.manta.joyent.com/rmustacc/public/tmp/connect.d >>>>> >>>>> The most useful thing to do would be to do the following: >>>>> >>>>> * Make sure drive is unplugged >>>>> * run the D script as: dtrace -qs connect.d -o /var/tmp/connect.out >>>>> * When you see the message in the system log about the connection on >>>>> port 12 failed, hit ctrl+c. >>>>> >>>>> If you can do this for each drive independently, that'd be appreciated. >>>>> If you can put each one of those files somewhere that'll help us figure >>>>> out what the next step is. >>>>> >>>>> Thanks, >>>>> Robert >>>> >>>> Here's one of the drives. I forgot to run the script before plugging in >>>> the second drive and unplugging/replugging didn't create any output. I >>>> can grab it the next time I'm able to reboot if it's needed. >>>> >>>> Syslog entries: >>>> >>>> 2017-01-06T02:28:27.152429+00:00 smarty usba: [ID 691482 kern.warning] >>>> WARNING: /pci@0,0/pci15d9,811@14 (xhci0): Connecting device on port 12 >>>> failed >>>> 2017-01-06T02:29:28.350770+00:00 smarty xhci: [ID 902155 kern.info] >>>> NOTICE: xhci0: xhci stop endpoint command (2)/slot (1) in wrong state: >>>> 19 >>>> 2017-01-06T02:29:28.350785+00:00 smarty xhci: [ID 617155 kern.info] >>>> NOTICE: xhci0: endpoint is in state 3#012 >>>> >>>> Full dtrace output here: http://pastebin.com/raw/rw6UDx2Q >>> >>> Thanks for the update Jason. Unfortunately it appears this is failing in >>> a different way than I really expected. Mainly it's getting to the point >>> that it seems like scsa2usb is attaching and starting to use the drive, >>> so the initial script that I provided isn't going to be the most >>> helpful. I'll try to put together something to further iterate on this. >>> >>> Thanks for all your help so far. >>> >>> Robert >> >> As a sanity check, I've been using both of these drives heavily for >> backups on USB 2.0 ports with no obvious errors. Monitoring SMART stats >> don't indicate there's a hardware issue that could impact these tests. >> >> Please let me know if I can provide any other information to help >> troubleshoot the new XHCI implementation. >> > > Hi Jason, > > Sorry for the delay in getting back to you. I think the next thing to > figure out is if scsa2usb is failing to attach. I suspect we're getting > far enough that it is. You can determine that by running: > > dtrace -n 'fbt::scsa2usb_attach:return{ trace(arg1); stack(); }' > > Would you mind opening a bug for this on > github.com/joyent/smartos-live/issues? I think that might make this a > bit easier to track.
Can you also include what the specific drives are that are failing to attach? Robert ------------------------------------------- smartos-discuss Archives: https://www.listbox.com/member/archive/184463/=now RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00 Modify Your Subscription: https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb Powered by Listbox: http://www.listbox.com
