[GTALUG] USB power reporting Type-c super speed skirmish and libpartd error

[Russell Reiter]

On Wed, Sep 23, 2020 at 2:27 PM D. Hugh Redelmeier via talk <talk at gtalug.org>
wrote:

> | From: Russell Reiter via talk <talk at gtalug.org>
>
> | On Wed, Sep 23, 2020 at 9:40 AM D. Hugh Redelmeier via talk <
> talk at gtalug.org>
> | wrote:
>
> | I've made some progress and then lost it again. I had to take a high def
> | photo of the writing on the cable to read that the cable is rated 3.0 and
> | my mb is only rated 3.0 but the peripheral card is usb 3.1. I think this
> is
> | why the mb connection hangs the system on usb 3.0, but the peripheral 3.1
> | card does not. I'm not sure which iteration of usb provides "alternative"
> | bus capabilities, ie. video and up to 100w power.
>
> You don't care about video.  We don't know if you need larger power
> for the disk.
>
> | The mb type-c 3.0 port reports overcurrent when the device is attached
> and
> | won't boot.
>
> Reports to who?  How?  What do *you* observe
>

udevadm monitor --env will report overcurrent events. It counted seven of
them at one point, at least it did until I reformatted to NTFS.

>
> | On the adapter card the type is usb 3.1. and the device doesn't
> | hang the system when booting,
>
> When booting from what?  From the NVMe card via USB C?
>

Sorry, I'm booting from a 250gib WD nvme in the primary m.2 slot on the
motherboard.

>
> |  but it also won't enumerate the drive
>
> What do you mean "enumerate"?  What do you meant by "it"?
>

I meant to say that the system wouldn't hang when plugged in but udevadm at
one point was endlessly binding and unbinding the device pci nodes without
transitioning to the usb mode alias.

>
>
> | >
> | > - reliably powering an NVMe drive, connected via USB C.
> | >   What are its specified power requirements?
>
> You didn't answer this question.
>

Sorry I don't have the spec's on the device immediately available, but I'm
going to look them up.

>
> | >   What can the ports provide?
> | >   USB C is a mess: one connector, several standards and options
> | >   within those standards, especially with respect to power.
> | >
> |
> | I may have to get a type-c 3.1 rated cable to go much further. It was all
> | working fine at one point before the power outage.
>
> I doubt that the cable is a problem.  But I guess that there's a
> chance.
>
> | I am updating the kernel
> | regularly on Fedora 31
>
> I think that that is a good idea but I think that it is unlikly to be
> the problem.
>

I'm going to reboot now, given that I was finally able to get the NTFS
partition recognized on the PCI-e cards Type-c port.

I don't necessarily hold much hope for consistency yet. I power off, unplug
the power cable and press and hold the start button for a few seconds to
see if I can dissipate any lingering bus voltage. I don't have a "manual"
troubleshooting direction to do that, but things did start acting better
when I started to do it. This was a practice I used to do diligently in the
90's, from before switched power supplies, but I never needed to do it much
after 2000.



> ---
> Post to this mailing list talk at gtalug.org
> Unsubscribe from this mailing list
> https://gtalug.org/mailman/listinfo/talk
>


-- 
Russell
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gtalug.org/pipermail/talk/attachments/20200923/56bf86f6/attachment.html>