FCNewInit and power management
by Callum Lerwick
I've been tweaking Fedora on laptops for a while now, and there really
is a huge need for a clean, sane, flexible userspace power management
infrastructure in Linux.
Everything is a scattered mess of duplicated effort and functionality.
Looking at it all, I've come to the realization that init/initscripts
really needs to be handling a lot of it! So I look around and discover
its been decided the whole SysvInit/initscripts system needs to be
rewritten anyway. Perfect! However looking at the wiki summary,
( http://fedoraproject.org/wiki/FCNewInit ) it doesn't seem as if anyone
else has realized the role init should have in power management.
Look at apmd's /etc/sysconfig/apm-scripts/apmscript, look at
acpid's /etc/acpi/actions/, look at suspend2's hibernate scripts
( http://www.suspend2.net/downloads/ ). Look at how init already (sorta)
handles UPS power failures! Is a desktop on a UPS really that much
different from a laptop with a battery? These are all extravagant
duplications of what run levels should be doing. (Except maybe acpid
which seems to do mostly nothing by default...)
Run levels are states the system can be in. In addition to the existing
"halt" "reboot" "single user" "no network" and "normal" states, there is
"on battery" and "suspended" states just begging to be added. (I'm not
sure the best way to handle slightly different types of similar states.
We could possibly have a "battery low" state and we have "standby"
"suspend to ram" and "suspend to disk" to choose from. Do they full run
levels of their own? I'm guessing not...)
The system I envision goes something like this. apmd and acpid get
stripped down so all they do is send events to init. init absorbs
similar functionality to acpid, it takes in "events" (over dbus?) from
apmd, acpid, UPS daemons, /usr/bin/poweroff, etc and evaluates rules
(Run scripts like acpid does?) that decide what states to switch to
based on these events. Switching states is then handled much like the
existing SysV init always has, or however the future system decides to
do so.
I'm actually not so sure about where the decision making needs to be,
but the main point is the state switching/tracking should all be done by
init, because thats what its for. Currently, there's a lot of
duplication in the handling of state changes, in apmd, acpid, hibernate,
laptop-mode, etc, that really needs to be centralized.
Comments?
18 years, 7 months
rawhide report: 20051018 changes
by Build System
Updated Packages:
NetworkManager-0.5.0-2
----------------------
* Mon Oct 17 2005 Christopher Aillon <caillon(a)redhat.com> - 0.5.0-2
- NetworkManager 0.5.0
dhcp-11:3.0.3-8
---------------
* Thu Oct 13 2005 Jason Vas Dias <jvdias(a)redhat.com> - 11:3.0.3-8
- further fix for bug 160655 / ISC bug 15293 - upstream patch:
do NOT always strip trailing nulls in the dhcpd server
evolution-data-server-1.4.1.1-1
-------------------------------
* Mon Oct 17 2005 David Malcolm <dmalcolm(a)redhat.com> - 1.4.1.1-1
- 1.4.1.1
* Mon Oct 17 2005 David Malcolm <dmalcolm(a)redhat.com> - 1.4.1-2
- Updated patch 102 (fix-implicit-function-declarations) to include fix for
http calendar backend (thanks to Peter Robinson)
* Tue Oct 04 2005 David Malcolm <dmalcolm(a)redhat.com> - 1.4.1-1
- 1.4.1
file-4.16-1
-----------
* Tue Oct 18 2005 Radek Vokal <rvokal(a)redhat.com> - 4.16-1
- upgrade to upstream
* Mon Oct 03 2005 Radek Vokal <rvokal(a)redhat.com> - 4.15-4
- file output for Berkeley DB gains Cracklib (#168917)
* Mon Sep 19 2005 Radek Vokal <rvokal(a)redhat.com> - 4.15-3
- small fix in previously added patch, now it works for multiple params
firstboot-1.3.51-1
------------------
* Mon Oct 17 2005 Chris Lumens <clumens(a)redhat.com> 1.3.51-1
- Fix whrandom deprecation warnings.
- Fix render_to_drawable deprecation warnings.
- Change "Next" button on last page to "Finish".
fontconfig-2.3.91.cvs20051017-1
-------------------------------
* Fri Oct 14 2005 Matthias Clasen <mclasen(a)redhat.com> - 2.3.91.cvs20051017-1
- Update to the mmap branch of fontconfig
gdm-1:2.8.0.4-6
---------------
* Mon Oct 17 2005 Steve Grubb <sgrubb(a)redhat.com> 1:2.8.0.4-6
- add login audit patch (bug 170569)
* Mon Oct 17 2005 Ray Strode <rstrode(a)redhat.com> 1:2.8.0.4-5
- bump redhat-artwork requirement to get rid of the boot
throbber for now, since it seems to have reappeared
mysteriously (bug 171025)
gjdoc-0.7.6-1
-------------
* Mon Oct 17 2005 Andrew Overholt <overholt(a)redhat.com> 0.7.6-1
- 0.7.6.
kernel-2.6.13-1.1616_FC5
------------------------
* Mon Oct 17 2005 Dave Jones <davej(a)redhat.com>
- 2.6.14-rc4-git5
libselinux-1.27.10-1
--------------------
* Mon Oct 17 2005 Dan Walsh <dwalsh(a)redhat.com> 1.27.10-1
-
lynx-2.8.5-24
-------------
* Mon Oct 17 2005 Tim Waugh <twaugh(a)redhat.com> 2.8.5-24
- Apply patch to fix CAN-2005-3120 (bug #170253).
mc-1:4.6.1a-0.19
----------------
* Sun Oct 16 2005 Jindrich Novy <jnovy(a)redhat.com> 4.6.1a-0.19
- update from CVS
- convert spec to UTF-8
- sync utf8, promptfix, 64bit patches
- drop upstreamed gcc4, ftpcrash, find, symcrash, cstrans, searchfix patches
- drop ctrl-t patch
- update userhost patch to let the edited/viewed file name be displayed in
xterm title
mkinitrd-5.0.5-1
----------------
* Mon Oct 17 2005 Peter Jones <pjones(a)redhat.com> - 5.0.5-1
- make PROBE, MODULES, and PREMODS load defaults from a config file
- get rid of support for not using a dynamic /dev
- add options to force device probes (for when PROBE is "no")
- consolidate some duplicate code paths
- make nash's otherCommand check PATH first and use what it finds there for
any commands it needs to run, unless the command starts with "nash-",
in which case the internal version is used unconditionally
- remove the symlink hack for setuproot/preswitchroot/switchroot
- wrap all the nash commands in mkinitrd with shell functions and use the
"nash-" varient.
- rework root fs creation so Jeremy can use the otherCommand feature
redhat-artwork-0.129-2
----------------------
* Mon Oct 17 2005 Ray Strode <rstrode(a)redhat.com> 0.129-2
- remove throbber patch for now
scim-anthy-0.7.1-1.fc5
----------------------
* Mon Oct 17 2005 Akira TAGOH <tagoh(a)redhat.com> - 0.7.1-1
- New upstream release.
- scim-0.7.0-fix_crash_bug.diff: removed.
selinux-policy-strict-1.27.1-18
-------------------------------
* Mon Oct 17 2005 Dan Walsh <dwalsh(a)redhat.com> 1.27.1-18
- Cleanup category defs
- Add configfs_t
- Allow http to run with mod_ftpd
- Fix initrc_exec_t scripts to run with the correct securitylevel
- Fixes for postfix_local
selinux-policy-targeted-1.27.1-18
---------------------------------
* Mon Oct 17 2005 Dan Walsh <dwalsh(a)redhat.com> 1.27.1-18
- Cleanup category defs
- Add configfs_t
- Allow http to run with mod_ftpd
- Fix initrc_exec_t scripts to run with the correct securitylevel
- Fixes for postfix_local
system-config-printer-0.6.144-1
-------------------------------
* Mon Oct 17 2005 Tim Waugh <twaugh(a)redhat.com> 0.6.144-1
- 0.6.144:
- Use SUDO_USER if set when printing test page (bug #170013).
- Avoid pam_stack (bug #170640).
util-linux-2.13-0.5.pre4
------------------------
* Mon Oct 17 2005 Karel Zak <kzak(a)redhat.com> 2.13-0.5.pre4
* fix #170564 - add audit message to login
vim-1:6.4.000-1
---------------
* Mon Oct 17 2005 Karsten Hopp <karsten(a)redhat.de> 6.4.000-1
- vim-6.4 patchlevel 0
wireless-tools-1:28-0.pre10.5
-----------------------------
* Mon Oct 17 2005 Christopher Aillon <caillon(a)redhat.com> 28-0pre10
- Update to version 28 pre10
yelp-2.12.1-3
-------------
* Mon Oct 17 2005 Matthias Clasen <mclasen(a)redhat.com> - 2.12.1-3
- Include the category General|Linux|Distributions|Other on the
title page
* Mon Oct 17 2005 Matthias Clasen <mclasen(a)redhat.com> - 2.12.1-2
- Fix a double-free bug
Broken deps for i386
----------------------------------------------------------
gdm - 1:2.8.0.4-6.i386 requires redhat-artwork >= 0:0.129-20
Broken deps for ppc
----------------------------------------------------------
gdm - 1:2.8.0.4-6.ppc requires redhat-artwork >= 0:0.129-20
Broken deps for s390x
----------------------------------------------------------
gdm - 1:2.8.0.4-6.s390x requires redhat-artwork >= 0:0.129-20
Broken deps for ia64
----------------------------------------------------------
gdm - 1:2.8.0.4-6.ia64 requires redhat-artwork >= 0:0.129-20
Broken deps for x86_64
----------------------------------------------------------
gdm - 1:2.8.0.4-6.x86_64 requires redhat-artwork >= 0:0.129-20
Broken deps for ppc64
----------------------------------------------------------
gdm - 1:2.8.0.4-6.ppc64 requires redhat-artwork >= 0:0.129-20
Broken deps for s390
----------------------------------------------------------
gdm - 1:2.8.0.4-6.s390 requires redhat-artwork >= 0:0.129-20
18 years, 7 months
Stateless Linux Update?
by Jake Thompson
Hi All,
I am wondering if there are any developers out there that can sound off on
the state of the stateless linux project. It would be a perfect solution for
what I am doing, but since it seems dead in the water, I am hesitant to move
forward.
Thanks,
Jake T.
18 years, 7 months
glibc/malloc_consolidate()
by Bojan Smojver
I have a feeling that recent versions of glibc in Rawhide have a problem
related to this function. It seems to be overwriting allocated memory.
Has anyone seen anything similar?
--
Bojan
18 years, 7 months
only half of screen is visable during boot
by Jason Dravet
I opened https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=170450 a week
and half ago and I have not received any response so I thought I would send
an email here. I have been using the rawhide kernel since
(2.6.13-1.1600_FC5 and I am running 1615 now. During the boot, only part of
the screen is visable. For example the monitor the is the lines and I can
only see text in the asterik area. The bottom of the area I can see is only
the top half of the line. It is like someone cut off the bottom of my
monitor.
-----------------------
| **************** |
| * * |
| * * |
| **************** |
| |
| |
| |
-----------------------
I have a Number Nine revolution 4 video card and a SGI 1600sw lcd panel.
Thanks,
Jason
18 years, 7 months
SATA DVD support
by ffedora
What is the status of SATA DVD support (Plextor PX-716SA) in Fedora 4?
18 years, 7 months
avahi and howl
by Trever L. Adams
I was just reading up on gaim development. They are moving to avahi in
place of howl. Debian is already doing or trying to do the same. It
seems avahi does have some functionality enhancements over howl and
license compability with more code (APSL vs. LGPL). It also supports
reflections over lan segments.
Is this something that RedHat/Fedora should follow Debian/Gaim on?
Sorry if this is purely political.
Thank you,
Trever Adams
18 years, 7 months
Silly request of the day:
by Stephen John Smoogen
With authconfig possibly getting a rewrite.. how about renaming it to
system-config-authorization ?
--
Stephen J Smoogen.
CSIRT/Linux System Administrator
18 years, 7 months