* Tue Apr 13 2004 Warren Togami <wtogami(a)redhat.com> 1:0.2.34-2
- remove INSTALL and 536k of useless .ps and html
Please let us know if you find any packages that contain large and very
not useful stuff like this. Very often we have no interest in the
"INSTALL" file since we already installed it via rpm, and in many cases
INSTALL is only the generic installation instructions. We can always
remove that safely. This is totally worth it if we can save a few MB of
space and reduce RPM file sizes.
Also any cases where development specific documentation (like API
specifications) are within the main package as %doc, we should move it
to the -devel package. This should save some space in desktop
On Thu, 15 Apr 2004 10:48:31 -0400
Matthew Miller <mattdm(a)mattdm.org> wrote:
> We're heavy AFS users here at Boston University, and it's going to be a
> major blow to not have an AFS client available. OpenAFS, last I looked, said
> that it'd take them a year to properly develop a 2.6 version *if* they got
> some funding to do so -- and they don't. Arla doesn't seem to be going
This is a fair assessement of the status, sans the "funding" whines.
It's not as if anyone else is rolling in money.
> So that leaves the in-kernel AFS -- which was apparently developed by
> someone at Red Hat. (An enterprise customer needed it, perhaps?)
The "someone"'s name is clearly present in the source, I believe,
and that is of David Howells. I and dwmw2 dabbed in it as well.
That client (kAFS) is not useful for any real work and I do not
see it becoming one before well into 2.7.
> Can someone point me to further information, or nicely tell me about, Red
> Hat / Fedora's plans in this regard? And what I can do to help? (From a
> tester's perspective at the very least.)
Feature requests come in, but David does have other responsibilites.
I try to do things about it in the community capacity, but it is not
going much of anywhere.
I think it would be the best if you found some grad students to
hack on AFS, either OpenAFS or kAFS. Both projects are amendable
to patches, in my experience, they just have no hacking cycles
to them. It's not important which one you decide to advance,
as long as you do.
If you cannot do that, well, that's tough. To stick to RHEL 3 and
OpenAFS would probably your best bet.
Usermode is the handy little program that makes all of the GUI
administration utilities (and some command line tools too -- it's not picky)
able to prompt for the root password when run as a normal user.
It also has a feature where users can, instead of authorizing as the root
user, authorize as themselves with their own password.
This is good for things like changing GECOS information, or anything else
where you want the user to demonstrate that they really are who they say
they are rather than just someone who walked up to the console.
However, it's on an all-or-nothing basis -- either everyone must give the
root password for a given program, or everyone can run it with their own
My patch implements what I call a "sudo-like" behavior (although it is much
simpler than sudo). Each program, through its console.apps config file, can
have a list of groups whose members are able to authorize as themselves.
Anyone not a member of the approved groups either must give the root
password (or the password of a given user, or is denied access completely
via a new <none> value).
This could allow members of an admin group (traditionally, 'wheel') to have
easy access to all of the administrative tools -- very reasonable for ease
of use on a personal desktop system. Or, you could be more fine-grained, and
give members of a certain group access to 'gtoaster' on a shared CD-burning
This all may sound complicated, but it's not. Usermode already implements
99% of what was needed -- the core patch is about a dozen lines! (There's
also is_group_member and is_grouplist_member helper functions, but those are
very simple too.)
And, it's a very non-evasive change, because if the config files aren't
changed, it defaults to acting exactly like it does now.
See the patch, and the request, at:
Any comments/suggestions are very welcome.
And Nalin or whoever else, please consider adding this.
Matthew Miller mattdm(a)mattdm.org <http://www.mattdm.org/>
Boston University Linux ------> <http://linux.bu.edu/>
Any reason why the fetchmail service debian & mandrake ship is not in
Fedora ? (fetchmail-daemon rpm in Mandrake)
It sure seems handy for people inside private networks (or when the
smtp port is filtered by the ISP).
On Fri, 2004-04-16 at 01:00, Warren Togami wrote:
> Also any cases where development specific documentation (like API
> specifications) are within the main package as %doc, we should move it
> to the -devel package. This should save some space in desktop
Just don't remove *useful* docs ;)
What about licenses, I've 362 copies of the GPL and LGPL in
/usr/share/doc/* now. Make use of symlinks ?
Nils Olav Selåsdal
w w w . u t e l s y s t e m s . c o m
I am having trouble accessing my FujiFilm FinePix 3800 camera in Raw
Hide (it worked perfectly in FC1 and earlier).
When I connect it and turn it on, logs get:
usb 1-1: new full speed USB device using address 8
SCSI subsystem initialized
Initializing USB Mass Storage driver...
scsi0 : SCSI emulation for USB Mass Storage devices
USB Mass Storage device found at 8
drivers/usb/core/usb.c: registered new driver usb-storage
USB Mass Storage support registered.
and the usb_storage and scsi_mod modules get loaded. However, the camera
does not seem to get "attached" to any specific /dev/xyz (or /udev/xyz
for that matter) - e.g. when I try to access /dev/sda, I get "No such
device or address".
% cat /proc/scsi/scsi
% cat /proc/scsi/usb-storage/0
Host scsi0: usb-storage
Product: USB Mass Storage
Serial Number: Y-406^^^^^030528XFPX0007040092
% cat /proc/bus/usb/devices
T: Bus=01 Lev=00 Prnt=00 Port=00 Cnt=00 Dev#= 1 Spd=12 MxCh= 2
B: Alloc= 0/900 us ( 0%), #Int= 0, #Iso= 0
D: Ver= 1.10 Cls=09(hub ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1
P: Vendor=0000 ProdID=0000 Rev= 2.06
S: Manufacturer=Linux 2.6.5-1.319custom uhci_hcd
S: Product=UHCI Host Controller
C:* #Ifs= 1 Cfg#= 1 Atr=40 MxPwr= 0mA
I: If#= 0 Alt= 0 #EPs= 1 Cls=09(hub ) Sub=00 Prot=00 Driver=hub
E: Ad=81(I) Atr=03(Int.) MxPS= 2 Ivl=255ms
T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 8 Spd=12 MxCh= 0
D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
P: Vendor=04cb ProdID=011a Rev=10.00
S: Product=USB Mass Storage
C:* #Ifs= 1 Cfg#= 1 Atr=c0 MxPwr= 0mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=08(stor.) Sub=05 Prot=00 Driver=usb-storage
E: Ad=81(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms
E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms
E: Ad=83(I) Atr=03(Int.) MxPS= 8 Ivl=1ms
I tried both 2.6.5-1.322 and 2.6.5-1.319custom (kernel-source recompiled
using stock config-i686 with 4G and 4G+4G disabled, APM disabled,
Am I doing something wrong? Or is this a bug (and should be reported)?
Is there some way to make it work? Thanks a lot for any help.
P.S. Other USB devices (printer, mouse) work OK.
Home Page: http://nogin.org/
E-Mail: nogin(a)cs.caltech.edu (office), aleksey(a)nogin.org (personal)
Office: Jorgensen 70, tel: (626) 395-2907
I've been working off and on for the last few weeks on a script for
generating a spec file for Perl modules from CPAN. The current
version is here:
I started noticing that source rpms generated by cpanflute2 needed a
lot of very redundant cleanup, although they would almost always
build. This script goes for the opposite approach. The spec file
generated is meant to be clean and correct (suitable for submitting to
fedora.us), since fixing build problems is usually fairly trivial.
This script currently does the following:
* Gets the name and version from the tar file name. (This probably
works for 99% of the modules on CPAN, but there are a few notable
* Parses any README files for something suitable for %description.
(I have a plan to make the guesses a bit better, but the current
code works on a lot of modules.)
* Guesses at what files should be included as %doc.
* Gets BuildRequires from PREREQ_PM in Makefile.PL.
* Sets URL and Source0 to actual (probably correct) URLs.
And there's probably other things I'm forgetting. :-)
Of course, the idea is to eventually either migrate these features
into cpanflute2 or obsolete it by moving its features into this
script. (I don't really care either way.)
To run the script, just provide the path to one or more tar files from
CPAN on the command line. I haven't implemented any usage message
(yet), so if you don't give it a list, it just won't do anything.
This message is all of the documentation that exists at the moment,
but if anyone other than me finds it useful, I'll fix that. :-)
Steven Pritchard - K&S Pritchard Enterprises, Inc.
Email: steve(a)kspei.com http://www.kspei.com/
Phone: (618)398-7360 Mobile: (618)567-7320
[root@ibmlaptop bin]# ls -l gftp*
-rwxr-xr-x 1 root root 278 Apr 14 21:34 gftp
-rwxr-xr-x 1 root root 301804 Apr 14 21:34 gftp-gtk
-rwxr-xr-x 1 root root 163828 Apr 14 21:34 gftp-text
I really want to just DELETE the gftp-text binary from the gftp package
in order to save some distribution package and filesystem space. gftp
the GUI client is great, but gftp-text really sucks. I can't imagine
that anybody prefers it over lftp or ncftp, and it has the usability of
Anybody see any good reason to keep gftp-text? If I don't hear any good
reason within the next 18 hours it will be deleted.