I finally tracked it down to some changes that were added to the sysctl to keep the OOM killer at bay.
The offending lines were:
vm.overcommit_memory=2
vm.overcommit_ratio=0
Bruce
On 9/19/07, Bruce Keats <brucekeats@xxxxxxxxx> wrote:
I am using the 2.6.22.5-76 kernel. When I try and use the 2.6.22.1-41 kernel, I get the same result even though the 2.6.22.1-41 kernel has worked on those machines in the past.I am using udev-113.12. I get the same result if I roll back to udev-113.11.Booting into single user mode also fails. It gets stuck in the same spot.Any ideas?Thanks,Bruce
On 9/19/07, Harald Hoyer <harald@xxxxxxxxxx> wrote:Bruce Keats wrote:
> Hi,
>
> With the latest set of updates within FC7, I are now seeing this message
> on several systems.
>
> udevd[581]: udev_event_run: fork of child failed: cannot allocate memory
>
> The message occurs at boot just after udev starts. The message is
> repeatedly constantly and there does not appear to be any way out.
> Control-C doesn't do anything. The message keeps repeating without any
> end.
>
> We have seen this on different H/W platforms. One is a dell laptop.
> One is an server with dual Xeon with 4 gig RAM. RAM tests on all the
> H/W checkouts OK. All H/W was running FC7 OK before. All systems are
> running the latest kernel with all the lastest updates.
>
> This seems to have started in the last couple of days or so. I can't
> seem to find any way around it.
>
> Is anyone else encountering the same problem? Any ideas how to track
> down the cause? Any ideas on how to fix it or get around it to the
> systems up?
>
> Thanks,
> Bruce
>
Hmm, first report. Haven't seen anything like that from other users. Was
there a kernel update in the last days?
--
fedora-list mailing list
fedora-list@xxxxxxxxxx
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
