rawhide report: 20060714 changes
Jim Cornette
fct-cornette at insight.rr.com
Sat Jul 15 01:04:44 UTC 2006
Paul Dickson wrote:
> On Fri, 14 Jul 2006 09:28:29 -0400, buildsys at redhat.com wrote:
>
>> kernel-2.6.17-1.2396.fc6
>> ------------------------
>> * Thu Jul 13 2006 Dave Jones <davej at redhat.com>
>> - 2.6.18-rc1-git7
>> - More lockdep fixes.
>> - Fix slab corruption issue.
>>
>> * Thu Jul 13 2006 Mike Christie <mchristi at redhat.com>
>> - Add iscsi update being sent upstream for 2.6.18-rc2
>>
>> * Thu Jul 13 2006 Roland McGrath <roland at redhat.com>
>> - Fix spec typo that swallowed kdump subpackage.
>
> This runs forever while attempting to install. I had to kill the nash
> process to get it to continue.
>
> $ ps faxu
> USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
> root 21145 2.6 10.7 149152 138424 pts/1 S+ 10:28 8:52 | \_ /usr/bin/python /usr/bin/yum update --exc
> root 30617 0.0 0.0 4444 1096 pts/1 S+ 11:19 0:00 | \_ /bin/sh /var/tmp/rpm-tmp.48603 11
> root 30622 0.0 0.0 4440 1188 pts/1 S+ 11:19 0:00 | \_ /bin/bash /sbin/new-kernel-pkg --
> root 31455 0.0 0.0 4440 496 pts/1 S+ 11:20 0:00 | \_ /bin/bash /sbin/new-kernel-pk
> root 31457 99.9 0.0 2516 792 pts/1 R+ 11:20 279:16 | \_ /sbin/nash --forcequiet
>
> I uninstalled it and tried again with the same results.
>
> -Paul
>
I encountered the same problem with the nash hanging. Instead of killing
nash, which consumed most of the cpu percentage and did not progress, I
killed pup which messed up the system with multiple versions. Nash was
still consuming a tremendous amount of cpu time. I then killed nash and
cleaned up the aftermath with rpm -Uvh --replacefiles --replacepkgs on
all but the kernel rpms. zlib had a scriptlet error but the other
non-kernel packages seem to be alright now. (used cached rpms from
/var/cache/development/packages)
Afterwards, I tried to do the same thing for the kernel with the -ivh
option instead and --replacefiles and --replacepkgs but rpm reported
defects within the kernel rpm. I did not download the kernel and try
again since the above repeated results from this message.
Jim
I uninstalled the last kernel and was going to
--
Performance:
A statement of the speed at which a computer system works. Or
rather, might work under certain circumstances. Or was rumored
to be working over in Jersey about a month ago.
More information about the test
mailing list