On 1/21/2013 10:52 AM, Mike Burns wrote:
On Tue, 2013-01-15 at 17:20 -0500, Jeff Bailey wrote:
> On 1/15/2013 3:31 AM, Nicolas Ecarnot wrote:
>> Le 14/01/2013 22:24, Jeff Bailey a écrit :
>>> On 1/14/2013 9:08 AM, Nicolas Ecarnot wrote:
>>>> Hi,
>>>>
>>>> I'm trying to install oVirt node 2.5.5-0.1.fc17 and it is node going
>>>> well. I'm trying to install it on a Dell blade M610 via a iDrac, and
>>>> I'm quit accustomed to this kind of thing.
>>>>
>>>> - As I already had the issue of the label on the grub boot line, I
>>>> knew how to specify /dev/sr0 or sr1 on the linux line. Well.
>>>> - The install is ok with discovering my hard drive, and the
>>>> installation seems to be fine. After rebooting, the bott process is
>>>> leading me to the login prompt where the documentation tells me to log
>>>> as 'admin' with the correct password :
>>>>
>>>> The keyboard is not responding, I can not even type the user name.
>>> Try adding usbcore.autosuspend=-1 to the kernel command line. I've had
>>> problems with iDRAC 7 virtual consoles with both Fedora 17 and 18 and
>>> this fixes it. I don't know about iDRAC 6 but trying it couldn't
hurt
>>> :) Getting the machine to always boot with that option could be a bit
>>> tricky with Node. I'm not sure if you'll just be able to edit the
>>> grub.conf and persist it or not.
>> Thank you very much, Jeff, this did the trick !
> You're welcome! I know that suddenly losing console access to your
> blades can be a bit distressing :)
>
>> And better than that : adding it on the first use in the kernel line
>> leads it to be permanently added in the grub conf even after a reboot.
>>
> That's very nice to know. I was afraid it would be tricky with Node but
> I guess it isn't.
>
>> Nice !
>>
Thanks for the update on this! Out of curiosity, was the need for
usbcore.autosuspend=-1 the result of some sort of bug? or is there a bug
filed for tracking the issue? If it's something that is going to be
needed long term, I'll consider adding it to the default kernel
arguments for ovirt-node.
After a *very* brief search I found
https://bugzilla.redhat.com/show_bug.cgi?id=850785 which isn't terribly
helpful. It does accurately describe the situation but it doesn't seem
to be getting much attention (all comments are from a single day).
Thanks
Mike
> _______________________________________________
> Users mailing list
> Users(a)ovirt.org
>
http://lists.ovirt.org/mailman/listinfo/users