On Thu, Nov 21, 2013 at 9:06 PM, Heinz Diehl <htd(a)fritha.org> wrote:
> What precisely are you asking for?
Just wanted to know if rpm is better than apt or vice-versa or both
are equal...or why did this bifurcation took place if the Linux is
only the kernel...?
On Thu, Nov 21, 2013 at 10:21 PM, Greg Woods <woods(a)ucar.edu> wrote:
> Probably a reference to the very early days of RPM (pre-yum). You'd
> install a package, then find some library was missing and go to install
> that, which led to something else missing, etc. A few cycles of this and
> you know what "dependency hell" means. Nowadays, with yum, all the
> dependencies are pulled in automatically, so "rpm hell" is largely a
> thing of the past.
Okay got it. I don't think saying it is worthy nowadays. But for an
end user this all won't matter anytime.
On Thu, Nov 21, 2013 at 9:08 PM, Dr. Michael J. Chudobiak
> yum didn't always exist. yum handles the dependencies between packages.
> Before that, a lot of prayer was involved if you mixed rpm packages from
> different sources.
You mean from third party sources.
claims openssh-6.3p1-4 and later, in conjuntion with openssl-1.0.1e-30.fc20
will allow to use ECDSA keys for pubkey authentication.
I have existing ECDSA keys from my non-Fedora installations using
vanilla openssh and openssl packages.
So I installed openssh-6.3p1-5 and openssl-1.0.1e-30. When trying to
use the aforementioned private ECDSA key to access one of the other
machines, it doesn't work. I'm getting an error message instead:
$ ssh some-machine
key_from_blob: EC_KEY_new_by_curve_name failed
Worse, I can't even connect to other machines using an RSA key, even if
I change the remote authorized_keys file accordingly. The debug output
shows that in this case the connection attempt goes more or less
through, until the connection stops with the same message as above:
key_from_blob: EC_KEY_new_by_curve_name failed
The openssh ECDSA key format can't be different between different
versions of openssh, otherwise the keys wouldn't be usable on different
I also didn't see any bug reports that the new openssh version would not
be able to use ECDSA keys for some reason (ssh-keygen allows to create
new ECDSA keys, after all). If I revert to the vanilla openssh/openssl
versions, connecting via ECDSA key works as expected again.
Does anybody have an idea what the problem could be?
Booting after today's kernel update, I noticed that there was a dkms
Nov 17 10:08:10 sds-desk-2 vboxdrv: Starting VirtualBox kernel
Nov 17 10:08:10 sds-desk-2 vboxdrv: (modprobe vboxdrv failed.
Please use '
dmesg' to find out why)
Nov 17 10:08:10 sds-desk-2 dkms_autoinstaller: Starting dkms:
Error! Could not locate dkms.conf file.
Nov 17 10:08:10 sds-desk-2 dkms_autoinstaller: File: does not exist.
# locate dkms.conf
Does anyone else using the Oracle distribution of VirtualBox see any
errors after a kernel update?
I recently installed Fedora 19 x86_64 on my previously running Fedora 16 x86_64.
I have been trying to figure out what happened to my audio on a USB head set. Finally I found a bug with symptoms but that was closed https://bugzilla.redhat.com/show_bug.cgi?id=424621.
I started looking at the bug and tried some of their commands. Not much there until I tried "pulseaudio -vvv"... Sound started working!
My question is, "Is this a Fedora bug or is there something that I need to configure to get pulseaudio started as a service (if this is a good idea)"
Regards and thanks to any who care to send hints/tips/suggestions/clues, etc.
[AO_ALSA] alsa-lib: pulse.c:243:(pulse_connect) PulseAudio: Unable to connect: Connection refused
My name is Darlene Wallach. I purchased a 2T Seagate FreeAgent GoFlex
Desk disk from Costco a few years ago. I finally set it up and found
your very immensely helpful directions!
Thank you very much!!!
On Thu, Sep 29, 2011 at 3:33 PM, linux guy <linuxguy123(a)gmail.com> wrote:
> Today I bought 2 3 TB hard drives to use in my new server.
> I'm writing this in case other people aren't aware that some of the
> traditional hard drive management methods don't work for drives larger than
> 2 TB.
> Specifically, fdisk and MSDOS type partition tables.
> Here is how I created a single partition on my 3 TB drives.
> Via parted.
> #parted -a optimal /dev/sdb
> (parted) rm 1 <- this removed the 2TB partition that fdisk made.
> NOTE: fdisk has a 2 TB limit, apparently forced by MSDOS style partition
> tables. Ever heard the story about how the width of a donkey ultimately
> determined the diameter of the space shuttle ?
> (parted) mklabel gpt <- this is NOT making a label for the drive. "man
> parted" is your friend.
> (parted) mkpart primary ext4 1 -1 <- this makes the partition itself, using
> the entire drive, which is what I wanted.
> (parted) print
> (parted) quit
> # mkfs -text4 /dev/sdb1 <- this formats the partition
> #e2label /dev/sdb1 myth <- this gives it a label, as used by mount in
> fstab, ie "LABEL=..."
> Via gparted.
> It turns out that the whole process can be done from within gparted, if you
> know what you are doing.
> The first thing you need to do is create the partition table.
> IF you have used fdisk on the drive previously, the partition table type is
> going to MSDOS, which has the 2TB limit. If you continue with the MSDOS
> partition table, gparted will allow you to select a 3TB partition to make,
> but it will give you the following error:
> partition length of 5860528002 sectors exceeds the
> msdos-partition-table-imposed maximum of 4294967295
> To get around this, select Advanced on the partition table options window
> and set the partition table type to gpt.
> Then you can proceed as you normally would to partition a drive with
> gparted. Ie add new partition(s), set the partition type, labels, etc.
> At this point you have an ext4 formatted 3TB drive.
> However, in F15 KDE at least, it doesn't seem to automount like a drive with
> an MSDOS partition table. The drive will not automatically appear in the
> Places column in Dolphin, for example. Nor will it appear in the list
> created by df.
> However, it is available to manually mount. Ie, #mount /dev/sdb1 temp,
> just like the old days.* And once you do that, it appears in the df list,
> though still not in the Places column in Dolphin.
> I hope this helps someone.
> *This might be particular to my machine being that the drive is a hot swap
> SATA device and I did not remove it and replace it. However, drives with
> MSDOS type partition tables auto mounted using the very same command
> users mailing list
> To unsubscribe or change subscription options:
> Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
equal justice under law
My system seems to have versions 2.0.19-5 and 2.0.20-2 both installed.
I thought that installing the later version would remove the earlier.
And why doesn't yum report a conflict? Is this a problem? Can I
simply remove the earlier version?
# rpm -qa acpi\*
# fgrep acpi yum.log
Feb 26 13:41:58 Updated: acpid-2.0.18-2.fc18.x86_64
Mar 06 08:34:26 Updated: acpid-2.0.18-3.fc18.x86_64
Jun 06 22:35:31 Updated: acpid-2.0.19-1.fc18.x86_64
Aug 24 19:43:22 Updated: acpid-2.0.19-5.fc19.x86_64
Aug 26 22:47:17 Updated: acpid-2.0.19-6.fc19.x86_64
Sep 24 22:34:40 Updated: acpid-2.0.20-1.fc19.x86_64
Sep 24 22:35:34 2:microcode_ctl-2.0-4.fc19.x86_64: ts_done name in te is
acpid should be 2:microcode_ctl-2.0-4.fc19.x86_64
Nov 16 10:43:29 Updated: acpid-2.0.20-2.fc19.x86_64
Nov 16 10:43:54 libdmapsharing-2.9.18-1.fc19.x86_64: ts_done name in te
is acpid should be libdmapsharing-2.9.18-1.fc19.x86_64
Thanks in advance - jon
i'm looking for a media player (DVD and divx) for my son's computer
using Fedora 19.
i need a particular feature as he's only 3 years old... i need to have
the same feature as under GOM Player... to tell the media player to
shutdown computer after the DVD/divx movie ended... like it exists in
GOM Player for Windows world.
i didn't find such feature in VLC or SMPlayer... and i really loved
SMPlayer from Fedora 17, too bad.
do you know some media player that integrate such feature ?