Anaconda installs OOM with selinux-policy-targeted rpm from F9 updates
by Martin Langhoff
One the last rounds of testing the new release of OLPCXS, I rebuilt it
with fresh packages from F9 update and started testing the installer.
Funny thing, the install did not complete -- instead, the machine
would switch off after spending a few minutes trying to install
selinux-policy-targeted.
After a few attempts to diagnose the problem, I managed to see vmstat
go all the way down to almost 0 memory just before the machine turned
itself off. This particular machine has ~980MB RAM available to the
OS. Tested on another machine with a proper 1GB memory, vmstat hit
bottom at ~10MB free while installing selinux-policy-targeted but
quickly recovered.
I know I've installed earlier F9 based spins on machines with 512 MB
of physical RAM so this seems like a fairly bad regression, specially
considering that I'll soon need to install this on a machine with
256MB RAM.
So I suspect we have 2 problems
- Selinux-policy-targeted instalation seems to have balooned into a
memory hog between f9-release and f9-updates
- Anaconda OOMs without a warning or useful message to the user
Has anyone else seen this? Diagnostics to recommend? I can
successfully log anything to disk until moments before the OOM
shutdown.
Will post version numbers of the selinux rpm tomorrow when I get back
in front of the offending machine and installer image. Apologies for
the vagueness :-)
cheers,
m
--
martin.langhoff(a)gmail.com
martin(a)laptop.org -- School Server Architect
- ask interesting questions
- don't get distracted with shiny stuff - working code first
- http://wiki.laptop.org/go/User:Martinlanghoff
15 years, 6 months
f8 packages not upgraded by f9
by Matthew Woehlke
While peeking at my list of orphaned packages, I noticed these, from f8:
grub-0.97-33.1.fc8.x86_64
hal-info-20080607-2.fc8.noarch
In the case of grub, it looks like the f9 distro tag isn't being
considered as newer than f8 when the version is otherwise the same (bug?).
For hal-info, it looks suspiciously like an f8 update postdated the f9
release and is versioned such that the f8 is "newer". I guess the
even-newer f10 package will just "fix" this, but it's still strange.
Any thoughts?
--
Matthew
Please do not quote my e-mail address unobfuscated in message bodies.
--
In the beginning, there were not enough colors. -- Guy Keren
15 years, 6 months
CC'ed to my own bugs
by Debarshi Ray
Since the Bugzilla update, I find that by default I am set to be CC'ed
to a bug which is already assigned to me. This has been like this for
quite some time now, so is this how it is meant to be?
Cheers,
Debarshi
15 years, 6 months
How to get an SELinux policy change
by Jerry James
I'm working on getting GCL to run again. The current Debian patch
(which is enormous) fixes most problems, but not the long-running
SELinux problem that GCL has had. I took a hint from a thread on this
list a couple of months ago. I let make run until it crashed due to a
denied mprotect() call, did chcon -t java_exec_t on the binaries, and
restarted the make. It completed successfully. I can patch the
makefile to do the chcon call in the right place, but I'm worried
about getting the right security context on installation now. First,
is using java_exec_t in this way acceptable? Second, if so, how do I
ask for Fedora's policy to reflect that: bugzilla, request on this
list, some other list? Thanks,
--
Jerry James
http://loganjerry.googlepages.com/
15 years, 6 months
rawhide report: 20081107 changes
by Fedora compose checker
Compose started at Fri Nov 7 06:01:25 UTC 2008
New package rcssbase
Robocup 2D Soccer Simulation Base Library
Updated Packages:
PackageKit-0.3.9-4.fc10
-----------------------
* Wed Nov 5 17:00:00 2008 Richard Hughes <rhughes(a)redhat.com> - 0.3.9-4
- Increase the timeout for cleaning up unused transactions. Due to a bug
in the PkClient library the new TID was not being requested, and the old
TID was being re-used. This gave a DBUS error if the user spent longer than
five seconds entering the password the very first time they used PackageKit
to do an authentication.
Apply a simple patch to mitigate this, as a more invasive (and correct)
patch is upstream. A new release will follow in f10-updates. Fixes rh#469950
cherokee-0.10.0-2.fc10
----------------------
* Thu Nov 6 17:00:00 2008 Tom "spot" Callaway <tcallawa(a)redhat.com> - 0.10.0-2
- do not package spawn-fcgi files (lighttpd-fastcgi provides them)
Resolves bz 469947
- get rid of rpath in compiled files
* Fri Oct 31 18:00:00 2008 Pavel Lisy <pavel.lisy(a)gmail.com> - 0.10.0-1
- updated to 0.10.0
cryptsetup-luks-1.0.6-6.fc10
----------------------------
* Thu Oct 30 18:00:00 2008 Milan Broz <mbroz(a)redhat.com> - 1.0.6-6
- Wipe old fs headers to not confuse blkid (#468062)
fedora-logos-10.0.1-2.fc10
--------------------------
* Thu Nov 6 17:00:00 2008 Tom "spot" Callaway <tcallawa(a)redhat.com> 10.0.1-2
- pull .git files out of source tarball to keep SRPM size down
* Thu Nov 6 17:00:00 2008 Tom "spot" Callaway <tcallawa(a)redhat.com> 10.0.1-1
- fix broken xfce4 icon (bz 470353)
- own directories for clean removal (bz 169282)
mock-0.9.13-1.fc10
------------------
* Thu Nov 6 17:00:00 2008 Jesse Keating <jkeating(a)redhat.com> - 0.9.13-1
- Add configs for F10 (jkeating)
torque-2.1.10-6.fc10
--------------------
* Wed Apr 16 18:00:00 2008 Garrick Staples <garrick(a)usc.edu> 2.1.10-6
- add alternatives system
Summary:
Added Packages: 1
Removed Packages: 0
Modified Packages: 6
Broken deps for ppc64
----------------------------------------------------------
livecd-tools-019-1.fc10.ppc64 requires yaboot
15 years, 6 months
Detecting binaries in rpmbuild
by Orion Poplawski
I just discovered in a package I'm putting through review that the
upstream tar ball contains some pre-compiled binaries. It seems like
this would be a good check for rpmbuild to run automatically before the
%build step. Thoughts?
--
Orion Poplawski
Technical Manager 303-415-9701 x222
NWRA/CoRA Division FAX: 303-415-9702
3380 Mitchell Lane orion(a)cora.nwra.com
Boulder, CO 80301 http://www.cora.nwra.com
15 years, 6 months
Re: [REMINDER] next monday cmake 2.6.2 will be required
by Matthew Woehlke
Alexander Neundorf wrote:
> in case you didn't update yet to cmake 2.6.2, please do so soon, next monday
> it will be required, as announced two weeks ago.
This is for KDE trunk. I noticed that 2.6.2 is already in koji. Any
plans on pushing it to updates (for F9) in the (very) near future?
--
Matthew
Please do not quote my e-mail address unobfuscated in message bodies.
--
In the beginning, there were not enough colors. -- Guy Keren
15 years, 6 months
firefox, mozplugger or evince: what messes up pdf's in firefox?
by sean darcy
On an F10 box updated often, for the past week, if I open a pdf url in
firefox, evince opens in sort of a separate window which is Always On
Top (for instance, any Print dialogs can't be seen since they're
"behind" the window). No window manager stuff ( minimize , close ). Only
way to close is from the File menu.
I'd file a bug, especially this close to a release, but whose bug is it?
sean
15 years, 6 months