FC2T2 rpm -qi unknown option
by Will H. Backman
[root@cei3440 root]# rpm -q kernel
kernel-2.6.3-2.1.253.2.1
[root@cei3440 root]# rpm -qi kernel
-qi: unknown option
[root@cei3440 root]# rpm -q -i kernel
-i: unknown option
Man pages still list it as a valid option.
--
Will Backman <whb(a)ceimaine.org>
Coastal Enterprises, Inc.
20 years, 1 month
Custom kernels
by Hans Engelen
Anyone notice btw (or is it old news) that the RHN-applet issues this warning
when running a custom kernel :
You are running kernel-xxxcustom, but it is not installed in RPM format. It is
highly recommended that you run the official Red Hat Linux kernels. By
running a custom kernel, you will miss out on any security fixes and
functionality enhancements provided by Red Hat Linux, and your configuration
is unsupported.
Must be an aprils fool joke right ?
I mean come on ... The great strenght of Linux is that you can configure and
customize the kernel options to your personal needs. Am I to understand from
this warning that Red Hat will refuse support to people that do so ? I can
understand not wanting to get blamed for instabilities that might arise from
kernels patched with unstable/untested 3rd party patches but customizing
your .config (and recompiling) on an otherwise unalterd kernel source tree
should be allowed no ?
Just my two cents worth.
20 years, 1 month
FC2T2 Errors after up2date kernel and up2date
by Will H. Backman
Was able to run up2date for the first time today.
New kernel was not installed, although policy was, causing panic on
boot. Per this list, hit "a" at grub and appended enforcing=0 to boot
options.
rpm -q kernel shows only one kernel installed, /var/spool/up2date only
contains header and no any downloaded packages.
The I run:
[root@cei3440 spool]# up2date --nosig -fu kernel
Traceback (most recent call last):
File "/usr/sbin/up2date", line 12, in ?
import rpm
ImportError: /usr/lib/python2.3/site-packages/rpmmodule.so: undefined
symbol: rpmdsBT
I guess manual download and update is next. Testing is fun. :) Better
than Zork.
--
Will Backman <whb(a)ceimaine.org>
Coastal Enterprises, Inc.
20 years, 1 month
RE: fedora-test-list Digest, Vol 1, Issue 729
by Troy Carpenter
The version I currently have is the one you listed (gtkam-0.1.10-2.1).
In either case, I still had to --nodeps gimp to get things to work.
Troy
-----Original Message-----
Date: Wed, 31 Mar 2004 19:18:17 +0200
From: Leonard den Ottolander <leonard(a)den.ottolander.nl>
Subject: Re: Today's dependency problems
To: For testers of Fedora Core development releases
<fedora-test-list(a)redhat.com>
Message-ID: <1080753497.4755.45.camel(a)athlon.localdomain>
Content-Type: text/plain
Hi Troy,
> Actually, the problem is that the new gimp package (gimp-2.0.0-5)
> deletes that library and replaces it:
>
> # rpm -ql gimp | grep libgimp
> /usr/lib/libgimp-2.0.so.0
> So the dependency is that libgimp-1.3.so.26 is being removed and will
> no longer be available, not that it is currently not available.
What version of gtkam is that? gtkam-0.1.10-2.1 only Requires gimp >=
1.2. That dependency should be satisfied by libgimp-2.0.so ...
Leonard.
--
mount -t life -o ro /dev/dna /genetic/research
20 years, 1 month
vnc.so causing xorg X server to segfault
by Paul Iadonisi
Title says it all. This was an FC2T1 system kept up to date with yum
and upgraded to FC2T2 with yum, including manually following the
procedure to enable selinux (I believe in relaxed mode). There's not
much info about the crash in /var/log/XFree86.0.log other than the
sig11, so I don't know if it's a general problem or selinux related.
Commenting out the 'Load vnc' line in my XF86Config prevents the sig11.
I will be rebooting with selinux=0 to see if that fixes it. Meanwhile,
has anyone who has done a clean install run into this with vnc.so?
--
-Paul Iadonisi
Senior System Administrator
Red Hat Certified Engineer / Local Linux Lobbyist
Ever see a penguin fly? -- Try Linux.
GPL all the way: Sell services, don't lease secrets
20 years, 1 month
Up2Date "size" column
by Sean Earp
Hello All-
For as long as I can remember, up2date has always put 0k in the
size field for every single package that it offers to update. If you
select multiple updates, it will tell you that it has maybe 7k to
download, when in fact you are downloading 80+ MB of packages. Is
there a technological reason for this incorrect representation? Is
there any purpose for having the size field displayed when the
information is completely erroneous? It has been a problem for so
long that I don't know if it should be thrown at bugzilla, or if I (we)
should just keep living with it. Any thoughts?
-Sean
GPG public key: <http://homepage.mac.com/smearp/seanpgp.asc>
20 years, 1 month
Synaptics touchpad
by Hans Engelen
Just wondering (haven't downloaded FC2T2 yet due to ISP download constraints,
blech), is the synaptics touchpad extended functionality (doubletap,
multifinger etc) restored in FC2T2 ? It was entered into bugzilla for FC2T1
and was a big nuisance for me too so.
Incase it was not, I fixed it on my FC2T1 setup with the 2.6.4 kernel update
from RHN and by installing the synaptics driver for XFree that lives at this
address : http://w1.894.telia.com/~u89404340/touchpad/
If you are wondering about the patches to the kernel done by Dimitry Torokhov,
those are present already in the 2.6.4-1.298 (and up?) kernel. All I had to
do was download the Xfree driver and follow the instructions on how to
install it and modify XF86config and hey presto. I am not sure when or where
the doubletap functionality was lost and if this is by design or not, but in
a way I am glad since this driver allows me to finetune many of the extended
capabilities of the touchpad to my personal liking (acceleration, doubletap
timelimits etc etc).
Anyways, it worked like a charm on my Dell C600 so I am happy.
20 years, 1 month
FC2T2 install report
by Jack Bowling
Summary - Flawless install... eventually. I'm impressed.
==================
Hardware listing -
Mainboard: ASUS CUBX, INTEL 440BX/PIIX4 chipset
BIOS: AWARD
CPU: Intel P3 Coppermine 700 MHz
Power Supply: Enermax Whisper EG465P-VE
SCSI Host Adaptor: Symbios Logic NCR53C810a
CD-ROM: Plextor CD-RW PX-W124TS; Sony CD-R CDU-415 (both SCSI)
SCSI Hard Drives: /dev/sda - IBM DALS-3540
/dev/sdb - IOMEGA ZIP 100
IDE Hard Drives: /dev/hda - Fujitsu MPC3043AT
/dev/hdc - Maxtor 6L040J2
/dev/hde - Maxtor 6E040L0
/dev/hdf - Maxtor 5T030H3
/dev/hdg - Maxtor 52049H4
/dev/hdh - Maxtor 6L040J2
USB 1.1 controller (built in): Intel Corp. 82371AB/EB/MB PIIX4 USB (rev 1)
USB 1.1 hub: Cicero 7-port SU-107, Alcor Micro chipset
USB 1.1 bus: /dev/sdh - SmartDisk CFUSB CF Reader
Logitech M-BD58 Optical Mouse
PCI USB 2.0 controller: NEC Corp rev65 chipset
USB 2.0 hub: Cicero 4-port USB 2.0, TI TUSB2040/2070 chipset
USB 2.0 bus: /dev/sdc,d,e,f - SanDisk ImageMate 6-in-1 CF Reader
/dev/sdg - Maxtor OneTouch external hard drive
Epson Perfection 2450 Photo Scanner
Creative Labs Sound Blaster Extigy
Video: ATI Radeon 7000/VE
Soundcard(2nd): Creative Labs SB Live!
Parallel printer: Epson Stylus Color 860
Network Devices: Linksys Network Everywhere Fast Ethernet NC100
An overloaded box built for hard drive space; obviously not a gamer. Will
be splitting this one up soon.
==================
Downloading binaries
- Did an overnight Bittorrent grab. Woke up to 4 shiny new iso's.
Install attempt #1
- Burned first iso. Already had FCT1 on /dev/hdc1 and planned on using
same partition for T2.
- Booted off CD and fine through the initial setup screens until it went to
grab the librpm.so when it crashed cuz it couldn't find the file. Silly
me. So I did an md5sum and found all 4 iso's were bad. Uh huh. So much
for Bittorrent infallibility.
- Another 2 BT downloads and the first 2 iso's were fine. But disc3 and 4
refused to match the published MD5s. So I checked with Downloader for X
from a web mirror which promptly told me that iso's 3 and 4 were just
fine, thank you. Elected to believe D4X and went ahead with the install.
Install attempt #2
- Having already shown myself that a CD install would have been fine with
good CDs, I elected to set up a hard drive install. Note to self: if your
target drive is mounted as /home then the target drive will be
jb/FC2-test2-binary-i386, not /home/jb....
- So I loop mounted the disc1 iso and copied the vmlinuz and initrd files
to my /boot partition. Set up a new stanza in grub.conf for the T2
installer and off we went. Kitchen sink install as always for a test.
- What can I say? Not one single hiccup. I went to work with it installing
packages and came back to the completion/reboot screen.
Firstboot
- The usual stuff with firstboot and then I was in. Can't say I notice any
difference with xorg server but then my lowly old Radeon is *not* going
to stress any of the video side. HOWEVER, I quickly learned the
consequences of electing to enable the selinux extensions during the
install. I tried to do the s-c-x configs and every one of them refused my
root password. Definitely a cool feature. I will have to read up on
selinux but I assume that there are some hoops I have to jump through to
give me admin privileges. This makes setting up a box and then locking it
down that much easier.
To do
- Turn off all the silly unneeded system services.
- Kill graphical boot.
- Play and get my bugzilla fingers loosened up.
Conundrum
- Why did disc3 and 4 MD5s not match even though they were fine on the install???
--
Jack Bowling
mailto: jbinpg(a)shaw.ca
20 years, 1 month
SATA errors linger in test2
by Rio Baan
I have an ASUS P4G8X-Deluxe using the Silicon image on-board SATA RAID.
On install, the two raid drives are detected as hde and hdg (why no hdf?!?)
and since the contents are in NTFS, obviously the installer can't read them.
Instead anaconda instructs me to WIPE them out removing ALL DATA! It's not
really a bug, but an installer threatening to kill my large RAID for no
apparent reason is a bit scary.
The same SATA boot problems are still happening... once again, the grey PC
boot screen comes up and dies when it gets to "Setting Hostname". So that's
where my testing on that machine ended.
Also, any idea why FC2 is giving me two kernels? I don't have a twin CPU
machine (I recently put a a 533Mhz P4 3.06Ghz into my box) so now I can't
see why I am getting an SMP kernel as an added bonus.
Also, I noticed some anomalies in the text version of the installer. After
completing installing test2 in text mode, I am told to hit Enter to reboot
the system. Hitting Enter just makes the cursor go down the screen....
forcing me to give the computer the three finger salute.
Also, I tried installing test2 on a really old Pentium I machine in text
mode. When it came time to swap CDs, the computer decided to reset itself
instead. When I retried the install with no extra packages (and no CD
change) install was OK on that machine.
_________________________________________________________________
Protect your PC - get McAfee.com VirusScan Online
http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963
20 years, 1 month