Multiple FC11 Update Failures - Kernel 2.6.30.9-96.fc11

[jrickman]

Has anyone experienced kernel update failures when updating to various
versions of kernel 2.6.30.9-96.fc11?

I have 3 different hardware platforms that all experience issues when
updating to different versions of 2.6.30.9-96.fc11. All seem to "halt"
when loading this kernel. Observation of Console output shows "drm"
console output and previous bootlogs from "known working kernels" say EXT4
should be the next console messages. When loading 2.6.30.9-96.fc11 I
never, repeat NEVER, see the EXT4 console messages and the machine never
finish booting. I have waited in some cases as long as 15 minutes just to
give the new kernel "a fair chance" to load.

The only recouce I have when this kernel loading failure occurs is to
Ctrl-Alt-Del, reboot, and select the older kernel from the GRUB menu.

Since I have to reboot the systems to get them working again, I don't know
of any way to capture console output when the loading of the new kernel
"stalls", so opening a Bugzilla case isn't worth the effort (no way to
help the developers if I can't provide any console output). I am open to
any "non-destructive" ideas that do not require any complicated hardware
setups to gather additional data to sort out this issue.

Platform #1
Hardware: Jetway J7F4K1G5S-LF
Memory: 1GB
Purpose: router
Current kernel (uname -r): 2.6.29.4-167.fc11.i686.PAE

Platform #2
Hardware: "unknown" Intel Atom 230 mobo in AcerNAS 340 chassis
Memory: 2GB
Purpose: file server
Current kernel (uname -r): 2.6.30.8-64.fc11.x86_64

Platform #3
Hardware: Asus M3A78 Pro
Memory: 8GB
Purpose: workstation and VMware host
Current kernel (uname -r): 2.6.29.4-167.fc11.x86_64

The only common thread I can find in all three systems is the following
filesystem format setups:

/boot runs EXT3
/ runs EXT4
/home runs EXT4
/var runs EXT4
/tmp runs EXT4



--
fedora-list mailing list
fedora-list@...
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines

[Sam Varshavchik]

jrickman@... writes:

> The only common thread I can find in all three systems is the following
> filesystem format setups:
>
> /boot runs EXT3
> / runs EXT4
> /home runs EXT4
> /var runs EXT4
> /tmp runs EXT4

Given that EXT4 is a relative newcomer, and you claim that your boot
progress halts at the point where you expect the kernel to issue some EXT4
status messages, this strongly suggests EXT4 borkage in this kernel build.

These kinds of things happen, from time to time. Your only realistic option
is to wait for the next kernel update, and hope that it gets fixed. You may
consider opening a bug in Bugzilla; that might help things. If there are
multiple reports in Bugzilla pointing the finger at ext4, this should catch
someone's attention.

You may also consider booting the most recent working kernel with the
"forcefsck" option, to force a fsck on all your ext4 partitions, in the
event that your fubarage got triggered by minor filesystem corruption, but
that's a long shot.





--
fedora-list mailing list
fedora-list@...
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines