udevsend: main: error sending message: Connection refused

Jim Gifford lfs at jg555.com
Sun Feb 5 12:42:12 PST 2006

John Gay wrote:
> On Friday 03 February 2006 18:37, Gay, John A wrote:
>>> I eventually found a post from Kay on the hotplug list, saying that
>>> from udev076 the initscript has to start udevd --daemon (this apparently
>>> used to happen through udevsend).
>>>  A patch follows which fixes the udev initscript in init.d on a built
>>> system, works for me with  I'm open to suggestions/advice on
>>> how to handle this for the book (probably, the problem will disappear
>>> when the new udev stuff goes in ?)  Do we need new bootscripts for the
>>> short term, or would a patch be better (the problem is only in clfs, lfs
>>> is still using an older udev) ?
>> This also caused a problem with my USB Stick. I've manually added this
>> fix to my box.
> Well, my work-week is over, so I can have another gander at my box.
> I'm running a dual Opteron and I followed the pure-64 path. Adding the patch 
> allows the new system to update /dev properly, but now I get many more 
> errors, which I think are related to the non-working hotplug, which I see is 
> coming out of the book anyway. Just looking for confermation that I'm correct 
> in thinking these will go away if I re-build without the hotplug stuff after 
> the book is updated. The errors are:
> udevd-event[XXXX]: run_program: exec of program '/sbin/udev_run_hotplugd' 
> failed
> udevd-event[XXXX]: run_program: exec of program '/sbin/udev_run_devd' failed
> Cheers,
> 	John Gay
You don't need those rules John. Use the package we have at 
http://ftp.jg555.com/udev, this has all the rules and everything needed 
for a 2.6.15+greg kh's patch or a 2.6.16-rc{1,2} series kernel.

jim at linuxfromscratch.org
lfs at jg555.com

LFS User # 2577
Registered Linux User # 299986

More information about the cross-lfs mailing list