rawhide report: 20061103 changes
by Build System
New package libgnomekbd
A keyboard configuration library
Updated Packages:
SDL-1.2.10-9
------------
* Thu Nov 02 2006 Thomas Woerner <twoerner(a)redhat.com> 1.2.10-9
- fixed arch order in SDL_config.h wrapper
anaconda-11.2-1
---------------
* Thu Nov 02 2006 Chris Lumens <clumens(a)redhat.com> - 11.2
- Add name resolution support for IPv6 AAAA and ip6.arpa records (dcantrell).
- Center anaconda window (#213350).
- Make sure to set a default on the physical extent combo (#212317).
- Fix updates on USB keys under kickstart installs.
- Set language support on text installs (pnasrat, #212511).
- Fix downloading kickstart files from nonstandard port numbers (#212622).
- Add more fonts (katzj, #207428).
- Fix localhost6 line (dcantrell, #211800, #210050).
- Disable keepalive so we don't run out of file handles (#212571).
- Display an error if we try to clear nonexistent hard drives (#212377).
- Fix lang-table typo (#212304).
- Fix file descriptor leak (katzj, #212191).
- Fix ZFCP config, kickstart handling, and module probing (katzj, #210094).
- Fix widget names on netconfig dialog (notting).
- Always bring up the network if specified in kickstart (#191424).
- Add Sinhala and Telugu (katzj, #207426, #207428).
- Fix iutil traceback (pnasrat, #211830).
- Continue if vname or vparm are NULL (dcantrell, #211502).
- Add fonts-telugu (katzj, #207428).
- Fix package installation progress bar size (katzj, #210531, #211526).
- Fix installation with CTC networking (katzj, #210195).
- Force swap formatting on ppc upgrades (pnasrat, #206523).
- Forget partitioning changes when going back in the UI (#211255).
- Don't specify a stdout or stderr for s390 shells (#210481).
- Fix window dragging jerkiness (krh).
- Add --noipv4 and --noipv6 (#208334).
- Correct --onbiosdisk handling (#210705).
- Set runlevel 3 by default on VNC installs (#211318).
- Fix download failures/retries (pnasrat, katzj, #211117).
- Don't use unicode line drawing characters on vt100-nav (katzj, #208374).
- Fix /tmp/netinfo parsing (dcantrell, #207991).
- Use virtualization instead of xen for the group name (katzj).
- Write out iscsi config to kickststart (katzj).
- Add qla2400 and qla2xxx to late sort module list (katzj, #210886).
- Use a better regex for finding ISO loopback mount points (#205133).
- Set up baseUrl correctly when given a list of URLs (#210877).
- Network config screen fixes (dcantrell).
- Tweak min/max swap numbers for low memory (#189490).
- Only run auditDaemon if not in test or rootpath mode (pjones).
- Fixes for supporting medium-less devices (Rez Kabir AT dell.com, #207331).
- yum API fixes (pnasrat).
- System language doesn't have to be in lang-table (#176538).
- Fix UI traceback (#210190).
* Mon Oct 09 2006 Jeremy Katz <katzj(a)redhat.com> - 11.1.0.110-1
- Fix SELinux contexts for iscsi
- Fix traceback if addrepos isn't shown (#209997)
- Fix traceback looking up hostnames (clumens, #209672)
- Fix split media (pnasrat)
- Fix network to be enabled after install on non-network installs
* Fri Oct 06 2006 Jeremy Katz <katzj(a)redhat.com> - 11.1.0.109-1
- Fix iscsi for toolchain changes and targets with multiple IPs
- Validate ips like 9.1.2.3 (dcantrel, #209654)
eclipse-1:3.2.1-12.fc7
----------------------
* Thu Nov 02 2006 Ben Konrath <bkonrath(a)redhat.com> 3.2.1-12
- Move doc plugins to %{_libdir}/eclipse/plugins because of html is being
generated differently on different arches.
- Fix multilib problem when there are two or more jars within a jar.
* Wed Nov 01 2006 Ben Konrath <bkonrath(a)redhat.com> 3.2.1-11
- Move copy-platform to %{_libdir}/eclipse
- Move the platform.source, icu4j, icu4j.source, help.webapp and
update.core.linux plugins to %{_libdir}/eclipse/plugins because these plugins
have platform specific content. Some of the platform specific content may be
a result of bugs in libgcj. These need to be investigated.
- Disable building the help indexes on all archs so that we have the same doc
plugins on all archs.
- Remove org.apache.ant_1.6.5/bin/runant.py to avoid multilib conflicts.
- Repack all the jars and the jars within those jars. This is needed
to make this package multilib compatible.
- Put SWT symlinks in %{_libdir}/eclipse instead of
%{_libdir}/eclipse/plugins.
* Wed Nov 01 2006 Andrew Overholt <overholt(a)redhat.com> 3.2.1-11
- Use equinox initializer instead of old patch to core.runtime.
- Run initializer *after* splitting install into arch-specific and
arch-independent locations.
- Move copy-platform to arch-specific location.
- Get rid of broken symlinks in tomcat plugin.
ekiga-2.0.3-3.fc7
-----------------
* Fri Nov 02 2007 Daniel Veillard <veillard(a)redhat.com> - 2.0.3-3
- Resolves: rhbz#201535
- fixes build-requires for opal-devel and pwlib-devel
evolution-data-server-1.9.1-4.fc7
---------------------------------
* Thu Nov 02 2006 Matthew Barnes <mbarnes(a)redhat.com> - 1.9.1-4.fc7
- Add patch for Gnome.org bug #369168, #369259, and #369261
(misc camel bugs reported by Hans Petter Jansson).
gdb-6.5-14.fc7
--------------
* Thu Nov 02 2006 Jan Kratochvil <jan.kratochvil(a)redhat.com> - 6.5-14
- Fix "??" resolving of symbols from (non-prelinked) debuginfo packages.
- Fix "??" resolving of symbols from overlapping functions (nanosleep(3)).
- Also disable testcase "checkpoint.exp" for a possible kernel Bug 207002.
- Provided (disabled during build) threading testsuite from BEA.
* Sat Oct 14 2006 Jan Kratochvil <jan.kratochvil(a)redhat.com> - 6.5-13
- Fix deadlock accessing last address space byte; for corrupted backtraces.
* Sun Oct 08 2006 Jan Kratochvil <jan.kratochvil(a)redhat.com> - 6.5-12
- Disabled IPv6 until its user visible syntax gets stable upstream.
gmp-4.1.4-10
------------
* Thu Nov 02 2006 Thomas Woerner <twoerner(a)redhat.com> 4.1.4-10
- fixed arch order in gmp.h and gmp-mparam.h wrapper for all architectures
* Thu Nov 02 2006 Joe Orton <jorton(a)redhat.com> 4.1.4-10
- include ppc64 header on ppc64 not ppc header
gnome-screensaver-2.17.1-2.fc7
------------------------------
* Thu Nov 02 2006 Ray Strode <rstrode(a)redhat.com> - 2.17.1-2
- temporarily revert smart card patches since they got
broke in the switch to 2.17.1 and people can't login (bug
212194)
gnucash-2.0.2-1.fc6
-------------------
* Wed Oct 11 2006 Bill Nottingham <notting(a)redhat.com> - 2.0.2-1
- update to 2.0.2
- update docs to 2.0.1
icu-3.6-9
---------
* Wed Oct 18 2006 Caolan McNamara <caolanm(a)redhat.com> - 3.6-9
- Resolves: rhbz#213648 extend prev/next to handle ZWJ
* Wed Oct 18 2006 Caolan McNamara <caolanm(a)redhat.com> - 3.6-8
- Resolves: rhbz213375 (icu.icu5488.assamese.patch)
irqbalance-1:1.13-6.fc7
-----------------------
* Thu Nov 02 2006 Neil Horman <nhorman(a)redhat.com> - 1.13-6
- bumping up MAX_INTERRUPTS to support xen kernels
- rediffing patch1 and patch3 to remove fuzz
* Tue Oct 17 2006 Neil Horman <nhorman(a)redhat.com> - 1.13-5
- Making oneshot mean oneshot always (bz 211178)
* Wed Sep 13 2006 Peter Jones <pjones(a)redhat.com> - 1.13-4
- Fix subsystem locking
kudzu-1.2.60-1
--------------
* Thu Nov 02 2006 Bill Nottingham <notting(a)redhat.com> - 1.2.60-1
- yet more SUBCHANNELS related fixing (#204803)
- rewrite vio probing (#213457)
m17n-db-1.3.3-34.fc7
--------------------
* Thu Nov 02 2006 Mayank Jain <majain(a)redhat.com>
- Added ZWNJ to hi-inscript/phonetic
openssh-4.3p2-11.fc7
--------------------
* Thu Nov 02 2006 Tomas Mraz <tmraz(a)redhat.com> - 4.3p2-11
- merge sshd initscript patches
- kill all ssh sessions when stop is called in halt or reboot runlevel
- remove -TERM option from killproc so we don't race on sshd restart
openssl-0.9.8b-9.fc7
--------------------
* Thu Nov 02 2006 Tomas Mraz <tmraz(a)redhat.com> 0.9.8b-9
- aliasing bug in engine loading, patch by IBM (#213216)
* Mon Oct 02 2006 Tomas Mraz <tmraz(a)redhat.com> 0.9.8b-8
- CVE-2006-2940 fix was incorrect (#208744)
* Mon Sep 25 2006 Tomas Mraz <tmraz(a)redhat.com> 0.9.8b-7
- fix CVE-2006-2937 - mishandled error on ASN.1 parsing (#207276)
- fix CVE-2006-2940 - parasitic public keys DoS (#207274)
- fix CVE-2006-3738 - buffer overflow in SSL_get_shared_ciphers (#206940)
- fix CVE-2006-4343 - sslv2 client DoS (#206940)
parted-1.7.1-18.fc7
-------------------
* Thu Nov 02 2006 David Cantrell <dcantrell(a)redhat.com> - 1.7.1-18
- Detect Apple_Boot partition types correctly (#204714)
php-5.1.6-4
-----------
* Tue Oct 31 2006 Joseph Orton <jorton(a)redhat.com> 5.1.6-4
- rebuild for curl soname bump
- add build fix for curl 7.16 API
tmpwatch-2.9.9-1
----------------
* Thu Nov 02 2006 Miloslav Trmac <mitr(a)redhat.com> - 2.9.9-1
- Exclude aquota.group and aquota.user files
Resolves: 206258
- Don't require the user-specified paths to be absolute
Broken deps for s390
----------------------------------------------------------
systemtap - 0.5.10-1.fc7.s390 requires kernel >= 0:2.6.9-11
systemtap-runtime - 0.5.10-1.fc7.s390 requires kernel >= 0:2.6.9-11
17 years, 5 months
Re: gstreamer and selinux issue
by Louis Garcia
On Fri, 2006-08-11 at 08:01 -0100, Paul Howarth wrote:
> On Thu, 2006-08-10 at 16:31 -0400, Louis Garcia II wrote:
> > On Thu, 2006-08-10 at 10:15 -0400, Daniel J Walsh wrote:
> > > On Wed, 2006-08-09 at 20:31 -0400, Louis Garcia II wrote:
> > > > On Wed, 2006-08-09 at 18:12 -0400, Louis Garcia II wrote:
> > > > > I was able to setup the pitfdll plugin for gstreamer and use the win32
> > > > > codecs under fc5 with selinux enabled. The pitfdll plugin needed to be
> > > > > marked textrel_shlib_t and the codecs under /usr/lib/win32 marked lib_t.
> > > > > > This worked for FC5 under selinux and FC6 with selinux disabled. But
> > > > > selinux under FC6 seems to have changed. Is their another lable I
> > > > > should use, how can I debug this?
> > > > >
> > > > > -Thanks
> > > >
> > > > This is what I get:
> > > >
> > > > Aug 9 19:12:34 soncomputer kernel: audit(1155165152.723:10): avc:
> > > > denied { execstack } for pid=9530 comm="totem"
> > > > scontext=user_u:system_r:unconfined_t:s0
> > > > tcontext=user_u:system_r:unconfined_t:s0 tclass=process
> > > >
> > > > -Louis
> > >
> > > you can turn on allow_execstack or change the context of totem to
> > unconfined_execmen_exec_t
> > > chcon -t unconfined_execmem_exec_t /usr/bin/totem
> >
> > if I turn on allow_execstack would that be for everything
>
> Yes.
>
> > or just for totem?
> > What would be the most secure of these two options?
>
> Just changing the context type of totem.
>
> Paul.
Ok, I chaged the context type of totem and now it's:
-rwxr-xr-x root root system_u:object_r:unconfined_execmem_exec_t /usr/bin/totem
This seems to fix my problem. However I get a slightly different message now:
Aug 11 15:09:41 soncomputer kernel: audit(1155323379.605:36): avc: denied { execheap } for pid=3094 comm="totem" scontext=user_u:system_r:unconfined_execmem_t:s0 tcontext=user_u:system_r:unconfined_execmem_t:s0 tclass=process
what does it mean?
-Louis
17 years, 6 months
aiglx discussions and issues
by Florin Andrei
What is the best mailing list to discuss aiglx issues - in general, but
also specific to FC6?
For aiglx-related issues, which package should be selected when
reporting a bug to bugzilla?
--
Florin Andrei
http://florin.myip.org/
17 years, 6 months
rawhide report: 20061102 changes
by Build System
Updated Packages:
at-3.1.10-5.fc7
---------------
* Wed Oct 25 2006 Marcela Maslanova <mmaslano(a)redhat.com> - 3.1.10-5
- daylight-saving
* Tue Oct 24 2006 Marcela Maslanova <mmaslano(a)redhat.com> - 3.1.10-3
- new version from upstream 3.1.10
* Wed Aug 23 2006 Marcela Maslanova <mmaslano(a)redhat.com> - 3.1.8-82.fc6
- #176486 don't fork option added (patch from Enrico Scholz)
bug-buddy-1:2.16.0-4.fc7
------------------------
* Wed Nov 01 2006 Alexander Larsson <alexl(a)redhat.com> - 1:2.16.0-4.fc7
- Add extra information to autogenerated bug reports
ccid-1.1.0-1
------------
* Thu Nov 02 2006 Bob Relyea <rrelyea(a)redhat.com> - 1.1.0-1
- Pickup ccid 1.1.0
* Thu Jul 20 2006 Florian La Roche <laroche(a)redhat.com> - 1.0.1-5
- require initscripts for post/postun
cryptsetup-luks-1.0.3-3.fc7
---------------------------
* Wed Nov 01 2006 Peter Jones <pjones(a)redhat.com> - 1.0.3-3
- Require newer libselinux (#213414)
dhcdbd-2.2-1.fc7
----------------
* Wed Nov 01 2006 David Cantrell <dcantrell(a)redhat.com> - 2.2-1
- Set chkconfig start order to 97 so dhcdbd starts before NetworkManager
evolution-data-server-1.9.1-3.fc7
---------------------------------
* Wed Nov 01 2006 Matthew Barnes <mbarnes(a)redhat.com> - 1.9.1-3.fc7
- Add patch for Gnome.org bug #353924 (category sorting).
gjdoc-0.7.7-12
--------------
* Wed Nov 01 2006 Andrew Overholt <overholt(a)redhat.com> - 0.7.7-12
- Guard against missing rebuild-gcj-db in post and postun.
- Add Requires(post,postun): java-gcj-compat to solve above.
- Resolves: #213529.
hplip-1.6.10-5.fc7
------------------
* Wed Nov 01 2006 Tim Waugh <twaugh(a)redhat.com> 1.6.10-5
- Fixed debugging patch.
* Wed Nov 01 2006 Tim Waugh <twaugh(a)redhat.com> 1.6.10-4
- Allow debugging of the SANE backend.
htmlview-3.0.0-15.fc7
---------------------
* Wed Nov 01 2006 Matthias Clasen <mclasen(a)redhat.com> - 3.0.0-15
- Ensure that launchmail always brings up the mail component
of evolution. (#212718)
libsepol-1.15.2-1
-----------------
* Wed Nov 01 2006 Dan Walsh <dwalsh(a)redhat.com> 1.15.2-1
- Upgrade to latest from NSA
* Merged fix from Karl MacMillan for a segfault when linking
non-MLS modules with users in them.
libuser-0.54.8-1
----------------
* Thu Nov 02 2006 Miloslav Trmac <mitr(a)redhat.com> - 0.54.8-1
- Add importing of HOME from default/useradd
Related: #204707
m17n-db-1.3.3-33.fc7
--------------------
* Wed Nov 01 2006 Mayank Jain <majain(a)redhat.com>
- Added 09CE mapped to z in as-inscript (213372)
* Wed Nov 01 2006 Mayank Jain <majain(a)redhat.com>
- Imported m17n-db-indic-0.4.29.tar.gz from RHEL-5 package, changes happened from .28 version are
- Added few more key combinations for ta-typewriter keymap - bug 209088
- Added ZWJ for hi-inscript and hi-phonetic keymaps - bug 211576
- Corrected kn-kgp and kn-inscript keymaps for keymapping of X - bug 209963
openoffice.org-1:2.0.4-5.7
--------------------------
* Wed Nov 01 2006 Caolan McNamara <caolanm(a)redhat.com> - 1:2.0.4-5.7
- Resolves: rhbz#213203 openoffice.org.2.0.4.oooXXXXX.i18npool.extendgrapheme.patch
- Resolves: rhbz#212988 try /tmp if /home fails, and abort with a message instead of segv
- Resolves: rhbz#213329 "." from numeric keypad not converted to "," in appropiate locales
- Resolves: (partial) rhbz#213371 openoffice.org.2.0.4.oooXXXXX.i18npool.extendgrapheme.patch
- Resolves: rhbz#212814 openoffice.org-2.0.4.ooo71076.dtrans.64bitdragdrop.patch
- add openoffice.org-2.0.4.ooo71039.svx.purevirtual.patch for bad pure virtual
- add openoffice.org-2.0.4.ooo71077.sc.purevirtual.patch for bad pure virtual
- rebuild for curl
pcsc-lite-1.3.2-1
-----------------
* Thu Nov 02 2006 Bob Relyea <rrelyea(a)redhat.com> - 1.3.2-1
- Pick up 1.3.2
quota-1:3.13-1.2.3.1.fc7
------------------------
* Wed Nov 01 2006 Steve Dickson <SteveD(a)RedHat.com> 1:3.13-1.2.3.2
- Added range checking on -p flag (bz 205145)
- Error message prints garbage characters (bz 201226)
shadow-utils-2:4.0.18.1-1.fc7
-----------------------------
* Wed Nov 01 2006 Peter Vrabec <pvrabec(a)redhat.com> 2:4.0.18.1-1
- upgrade
tmpwatch-2.9.8-1
----------------
* Wed Nov 01 2006 Miloslav Trmac <mitr(a)redhat.com> - 2.9.8-1
- Add optional unit suffix to the "hours" (now "time") parameter (original
patch by Alan J Rosenthal <flaps(a)dgp.toronto.edu>)
- Fix some format string vs. arguments mismatches
- Fix some rpmlint warnings
vorbis-tools-1:1.1.1-4.fc7
--------------------------
* Wed Nov 01 2006 Matthias Clasen <mclasen(a)redhat.com> - 1:1.1.1-4
- Rebuild against new curl
- Don't use CURLOPT_MUTE
wireshark-0.99.4-1
------------------
* Wed Nov 01 2006 Radek Vokál <rvokal(a)redhat.com> 0.99.4-1
- upgrade to 0.99.4 final
* Tue Oct 31 2006 Radek Vokál <rvokal(a)redhat.com> 0.99.4-0.pre2
- upgrade to 0.99.4pre2
Broken deps for ppc64
----------------------------------------------------------
php - 5.1.6-3.ppc64 requires libcurl.so.3()(64bit)
php-cli - 5.1.6-3.ppc64 requires libcurl.so.3()(64bit)
Broken deps for i386
----------------------------------------------------------
php - 5.1.6-3.i386 requires libcurl.so.3
php-cli - 5.1.6-3.i386 requires libcurl.so.3
Broken deps for x86_64
----------------------------------------------------------
php - 5.1.6-3.x86_64 requires libcurl.so.3()(64bit)
php-cli - 5.1.6-3.x86_64 requires libcurl.so.3()(64bit)
Broken deps for ia64
----------------------------------------------------------
php - 5.1.6-3.ia64 requires libcurl.so.3()(64bit)
php-cli - 5.1.6-3.ia64 requires libcurl.so.3()(64bit)
Broken deps for s390
----------------------------------------------------------
php - 5.1.6-3.s390 requires libcurl.so.3
php-cli - 5.1.6-3.s390 requires libcurl.so.3
systemtap - 0.5.10-1.fc7.s390 requires kernel >= 0:2.6.9-11
systemtap-runtime - 0.5.10-1.fc7.s390 requires kernel >= 0:2.6.9-11
Broken deps for ppc
----------------------------------------------------------
php - 5.1.6-3.ppc requires libcurl.so.3
php-cli - 5.1.6-3.ppc requires libcurl.so.3
Broken deps for s390x
----------------------------------------------------------
php - 5.1.6-3.s390x requires libcurl.so.3()(64bit)
php-cli - 5.1.6-3.s390x requires libcurl.so.3()(64bit)
17 years, 6 months
FC6 - Stock x86_64 stock install, installs about 158 i386/686 rpm's
by Chris Stankaitis
Greetings Fedora Dev's
I grabbed fc6-x86_64 and did a stock install (with office/prod selected)
and to my horror the install included about 158 i386/686 RPM's. with a
big massive rpm -e line I was able to take them out. I had i386/686
duplicates for most programs, gaim/firefox/openssl etc...
To the best of my understanding every package for fedora x86_64 should
be included out of the box for x86_64, and that 32bit compatibility
should only be added after the fact if it's required for a 3rd party
apps not maintained by fedora?
I would like to keep my x86_64 box "clean" of 32 bit applications if
possible, and having to rpm -e 100+ of them as my first step of a
post-kickstart is a pain in the butt. has anyone else seen this? I am
willing to help track things down off the list if anyone would like to
investigate this further.
--Chris
17 years, 6 months
xine-lib xine-ui gxine
by Gertjan Vinkesteijn
I have the Subject field programs compiled and working on 64-bits
Fedora6, what shall we do with it?
gertjan@gershwin ~]$ gpg --finger gertjanvinkje
pub 1024D/3FC1E76E 2006-10-29 [expires: 2007-10-29]
Key fingerprint = 4E9F 82A6 8303 0D6E 40CC B41A E52C 22AE 3FC1 E76E
uid GerardJan Vinkesteijn (New! FedDocProj)
<gertjanvinkje(a)yahoo.es>
sub 1024g/BC77793E 2006-10-29 [expires: 2007-10-29]
______________________________________________
LLama Gratis a cualquier PC del Mundo.
Llamadas a fijos y móviles desde 1 céntimo por minuto.
http://es.voice.yahoo.com
17 years, 6 months
package updater optimization proposal
by Gianluca Cecchi
Hello,
I'm using package updater with my brand new fc6.
I found in the past in fc5 that using yum from command line, I could
do a Ctrl+C when hitting a slow mirror and then I got a faster one,
eventually with more than one Ctrl+C.
it was a useful workaround..
Only drawback was that the Ctrl+C was not trapped and so after the
downloading phase completed, the program exited and I had to rerun
"yum update" to go to the transaction phase.
In fc6 it seems Ctrl+C is trapped, so it seems better in this way.
In graphical package updater instead, you have no way to say to it for
example "hey, I got a slooooow mirror; can you change it, please? Also
randomly, I don't want to choose a specific one..."
Is it reasonable to have pup propose to change mirror through a button
or something else?
Or by deign is intended that the mirror you get, the mirror you deserve?
I admit that in this way you could saturate some mirrors, but during
these days using pup is a pain... I always get very slow mirrors. I'm
based in Italy and I'm using adsl with Tele2, that I think uses
servers based in sweden to go through broad internet...
Gianluca
17 years, 6 months
Re: vncserver in future dists
by Chris Tyler
Adam Tkac <atkac(a)redhat.com> wrote:
> Hi all,
>
> I'm going to port x11vnc
> (http://www.karlrunge.com/x11vnc/) project as default vncserver in
> future dists. Current vncserver has self xorg-server and this is
> argument for x11vnc. When main xorg-x11-server package is patched,
> patches must be ported to vnc now. vncserver might be only "layer" on
> xorg-x11-server and have dependency for it. Maintaining of vnc could be really easy.. What do you think
> about it?? Is this bad idea??
FC6 includes four VNC servers:
- the VNC module for the Xorg server (in package vnc-server)
- Xvnc (vnc-server)
- krfb (kdenetwork)
- vino-server (vino)
krfb and vino-server are basically KDE/Gnome implementations of the same
thing; Xvnc is headless; and the VNC module is a (potentially)
higher-performance version of krfb/vinoserver that is running whenever
the X server is running (i.e., you can login to a session with it).
x11vnc is similar to krfb/vino-server in that it's an X client and does
not contain the X server. Others have pointed out that it can be used
with an X server driving a dummy framebuffer to act like Xvnc, but that
seems kludgey.
One important feature of Xvnc that x11vnc seems to be missing is the
"-inetd" mode, where the RFB protocol is handled on stdin/stdout and the
X display number is automatically determined (first unused port in the
range 6000-6099). This feature is important not only for on-demand vnc
server instances but also for exotic applications such as X clients
embedded into web pages (e.g., embedding one specific X app into a
rectangle on a web page, used for product demos and so forth). One
x11vnc FAQ entry mentions using x11vnc with xinetd, but it's talking
about connecting to a running X server on demand, not starting a new X
instance; you can bridge the gap with some scripting, but again it seems
inelegant.
--
Chris Tyler
17 years, 6 months
NetworkManager not working in rawhide
by David Cantrell
Some people may have noticed that NetworkManager is not working
correctly in rawhide post-FC6. The problem is caused by dhcdbd not
starting before NetworkManager, so NM gets confused and starts running
around crazylike.
The dhcdbd-2.2-1.fc7 package fixes the daemon start order. It will
appear in tomorrow's rawhide. Just wanted to let everyone know.
--
David Cantrell
Red Hat / Westford, MA
17 years, 6 months