Re: grub2
by Patrick Dupre
> ----- Original Message -----
> From: Chris Murphy
> Sent: 05/05/14 07:35 AM
> To: Community support for Fedora users
> Subject: Re: grub2
>
> On May 4, 2014, at 4:33 PM, Patrick Dupre <pdupre(a)gmx.com> wrote:
>
> > Hello,
> >
> > After the installation of a new kernel (typically during an update).
> > There is a change of the file /boot/grub2/grub.cfg.
> >
> > However, the new menuentry created is not correct.
> > The new one includes
> > linux /vmlinuz-3.14.2-200.fc20.x86_64 root=/dev/mapper/VolGrpSys_DK1-root
> > which wrong.
> > I can change it manually.
> >
> > I cannot guess where the information is found!
> > the
> > set root='lvm/VolGrpSys_DK1-root'
> >
> > is correct
>
> The first one looks correct, the second one doesn't. Can you either post the entire grub.cfg, or you can also edit an older entry and check what root= it uses. For any rootfs on LVM, I expect root=/dev/mapper/VG-LV. When updating the kernel, it calls grubby, which should just create a new entry based on a previous entry only updating the name of the kernel and initramfs accordingly. I haven't seen it change the root entry.
>
> The other thing you can do is create a new grub.cfg from scratch by using
>
> grub2-mkconfig -o /boot/grub2/grub.cfg
>
> This is the same command the installer used when Fedora was installed.
>
Thank you for the information.
When I run grub2-mkconfig -o /boot/grub2/grub.cfg.test
the generated file is correct.
I am not able to reproduce manually what happens during the 2 last kernal updates.
I will check again after the next update.
===========================================================================
Patrick DUPRÉ | | email: pdupre(a)gmx.com
Laboratoire de Physico-Chimie de l'Atmosphère | |
Université du Littoral-Côte d'Opale | |
Tel. (33)-(0)3 28 23 76 12 | | Fax: 03 28 65 82 44
189A, avenue Maurice Schumann | | 59140 Dunkerque, France
===========================================================================
10 years
What on earth just happened?
by Tom Horsley
I ran yum update this morning, getting a gazillion updates.
I rebooted the system.
When it came back up, I couldn't log in. I'd get a KDM screen,
login, then immediately get the KDM screen again.
I tried to login with a "failsafe" session which gave me
(apparently) an xterm that printed black rectangles
for all characters :-).
I tried other sessions like xfce and gnome and got the
same KDM login screen popping back up after I typed
my password.
I got in a vterm as root, created a new user, and
logged in as that user with a default gnome session.
That seemed to work fine. I decided I should compare
what is running in the session with my original
user, so I logged out and logged in again as original
user, and now my original user login works fine.
Logging in once as a different user appears to have
completely fixed the system.
What the heck was that? (It will be interesting to
see if the same sequence happens when I install the
latest updates at home).
10 years
kernel/nfsd lockup
by Gary Artim
anyone have system crashes on NFS using current kernel? 2 nites in a
row I've crashed, backed out the last 2 kernels.
May 7 01:32:26 r1epi kernel: BUG: soft lockup - CPU#2 stuck for 22s! [nfsd:674]
May 7 01:32:26 r1epi kernel: Modules linked in: binfmt_misc sch_sfq
xt_iprange bonding ip6t_REJECT nf_conntrack_ipv6 nf_conntrack_ipv4
nf_defrag_ipv6 nf_defrag_ipv4 xt_conntrack nf_conntrack
ip6table_filter ip6_tables snd_hda_codec_hdmi snd_hda_codec_realtek
snd_hda_codec_generic iTCO_wdt iTCO_vendor_support btrfs
x86_pkg_temp_thermal coretemp eeepc_wmi asus_wmi sparse_keymap mxm_wmi
kvm crct10dif_pclmul crc32_pclmul crc32c_intel raid6_pq
ghash_clmulni_intel btusb bluetooth microcode xor serio_raw
6lowpan_iphc i2c_i801 rfkill e1000 lpc_ich mfd_core snd_hda_intel
snd_hda_codec snd_hwdep snd_pcm e1000e snd_timer snd mei_me ptp
soundcore mei pps_core shpchp wmi nfsd auth_rpcgss nfs_acl lockd
sunrpc i915 firewire_ohci firewire_core i2c_algo_bit crc_itu_t
drm_kms_helper drm video aacraid i2c_core
May 7 01:32:26 r1epi kernel: CPU: 2 PID: 674 Comm: nfsd Not tainted
3.14.2-200.fc20.x86_64 #1
May 7 01:32:26 r1epi kernel: Hardware name: System manufacturer
System Product Name/P8Z68-V PRO, BIOS 0902 09/19/2011
May 7 01:32:26 r1epi kernel: task: ffff880409dfc280 ti:
ffff88040933e000 task.ti: ffff88040933e000
May 7 01:32:26 r1epi kernel: RIP: 0010:[<ffffffff81202607>]
[<ffffffff81202607>] d_obtain_alias+0x1b7/0x1d0
May 7 01:32:26 r1epi kernel: RSP: 0018:ffff88040933fb00 EFLAGS: 00000202
May 7 01:32:26 r1epi kernel: RAX: ffff8804159b5800 RBX:
ffffffff812039a3 RCX: ffff8802a014ff09
May 7 01:32:26 r1epi kernel: RDX: ffff8804159b58b0 RSI:
ffffffffa06a1140 RDI: ffff88011500a298
May 7 01:32:26 r1epi kernel: RBP: ffff88040933fb18 R08:
0000000000017bf0 R09: ffffffff812021b5
May 7 01:32:26 r1epi kernel: R10: ff70d95e69a37803 R11:
ffffea001059dc00 R12: 0000000000000000
May 7 01:32:26 r1epi kernel: R13: ffff88007126bd40 R14:
ffffffff810d1cb5 R15: ffff88040933fa78
May 7 01:32:26 r1epi kernel: FS: 0000000000000000(0000)
GS:ffff88042fa80000(0000) knlGS:0000000000000000
May 7 01:32:26 r1epi kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
May 7 01:32:26 r1epi kernel: CR2: 00007fce32b47cc0 CR3:
0000000001c0c000 CR4: 00000000000407e0
May 7 01:32:26 r1epi kernel: Stack:
May 7 01:32:26 r1epi kernel: ffff88007126bd40 ffffffffffffff8c
000000000004027d ffff88040933fb88
May 7 01:32:26 r1epi kernel: ffffffffa0661375 00000000008fb47b
ffff880417fcc000 0000000009dfc280
May 7 01:32:26 r1epi kernel: 7bff880409dfc790 0100000000008fb4
0000000000000000 00000000f09fdf25
May 7 01:32:26 r1epi kernel: Call Trace:
May 7 01:32:26 r1epi kernel: [<ffffffffa0661375>]
btrfs_get_dentry+0x115/0x140 [btrfs]
May 7 01:32:26 r1epi kernel: [<ffffffffa01eaa40>] ?
nfsd_proc_getattr+0xa0/0xa0 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffffa0661662>]
btrfs_fh_to_dentry+0x32/0x60 [btrfs]
May 7 01:32:26 r1epi kernel: [<ffffffff812cc842>] exportfs_decode_fh+0x72/0x2e0
May 7 01:32:26 r1epi kernel: [<ffffffffa01f0a2b>] ? exp_find+0x10b/0x1c0 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffff810c2885>] ? sched_clock_cpu+0x85/0xc0
May 7 01:32:26 r1epi kernel: [<ffffffff811cc925>] ? kmem_cache_alloc+0x35/0x1f0
May 7 01:32:26 r1epi kernel: [<ffffffff810b3606>] ? prepare_creds+0x26/0x1c0
May 7 01:32:26 r1epi kernel: [<ffffffffa01eb796>] fh_verify+0x316/0x600 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffffa01bd10e>] ?
cache_check+0x2be/0x380 [sunrpc]
May 7 01:32:26 r1epi kernel: [<ffffffffa01f9ac9>] nfsd4_putfh+0x49/0x50 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffffa01fbd1a>]
nfsd4_proc_compound+0x56a/0x7b0 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffffa01e7dbb>]
nfsd_dispatch+0xbb/0x200 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffffa01b2d00>]
svc_process_common+0x480/0x6f0 [sunrpc]
May 7 01:32:26 r1epi kernel: [<ffffffffa01b3077>]
svc_process+0x107/0x170 [sunrpc]
May 7 01:32:26 r1epi kernel: [<ffffffffa01e774f>] nfsd+0xbf/0x130 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffffa01e7690>] ?
nfsd_destroy+0x80/0x80 [nfsd]
May 7 01:32:26 r1epi kernel: [<ffffffff810ae211>] kthread+0xe1/0x100
May 7 01:32:26 r1epi kernel: [<ffffffffa03509e8>] ?
bond_remove_proc_entry+0x58/0x80 [bonding]
May 7 01:32:26 r1epi kernel: [<ffffffff810ae130>] ?
insert_kthread_work+0x40/0x40
May 7 01:32:26 r1epi kernel: [<ffffffff816fef7c>] ret_from_fork+0x7c/0xb0
May 7 01:32:26 r1epi kernel: [<ffffffff810ae130>] ?
insert_kthread_work+0x40/0x40
May 7 01:32:26 r1epi kernel: Code: 00 66 41 83 45 58 01 66 83 83 88
00 00 00 01 48 89 de 4c 89 ef e8 ca 81 0e 00 4c 89 e8 e9 a1 fe ff ff
f3 90 48 8b 88 b0 00 00 00 <80> e1 01 75 f2 e9 75 ff ff ff 48 89 f8 e9
86 fe ff ff 0f 0b 0f
10 years
nsd service file bug
by David Benfell
Hi all,
I have reported a bug for nsd at https://bugzilla.redhat.com/show_bug.cgi?
id=1096049 for version 3.2.17-1.fc20
nsd does not support a -D option and therefore fails at start. A workaround
is to copy the service file to /etc/systemd/system, remove the -D option
from the ExecStart line, and change the service type from 'simple' to
'forking'. Then disable the service and re-enable it.
This is a bug that should not have happened and should not have been
released into the wild.
--
David Benfell
See https://parts-unknown.org/node/2 if you do not understand the
attachment.
10 years
Video editing
by Matthew Saltzman
I shot a video with my phone, but I got it upside down. Not
surprisingly, I guess, the usual video playback tools don't have an
option to invert the image. Is there anything in Fedora that can carry
out that sort of transformation?
TIA.
--
Matthew Saltzman
Clemson University Math Sciences
mjs AT clemson DOT edu
10 years
KDE using sudo instead of su
by Emmett Culley
I've noticed recently that when I run yumex, or most any other app that needs root permission, from the launcher menu, I am prompted for my user password, instead of the root password. So I assume sudo is being used instead of su.
The "run backend with sudo" is not checked in yumex preferences. If I set the menu item to execute as root and add --root to the yumex call I get the root pasword request but then yumex doesn't start.
It is interesting that my other workstation works differently. That is, yumex doesn't prompt for the root password until it is about to do something that requires root permissison. But it is asking for the root password, not my user password, as are all other apps that require root permissions.
Any ideas how to get apps to require the root password instead of my user password?
Emmett
10 years
Which fedora?
by Patrick Dupre
Hello,
On a machine with only 1 Go of RAM, which version of fedora would you
recommand?
Thank.
===========================================================================
Patrick DUPRÉ | | email: pdupre(a)gmx.com
Laboratoire de Physico-Chimie de l'Atmosphère | |
Université du Littoral-Côte d'Opale | |
Tel. (33)-(0)3 28 23 76 12 | | Fax: 03 28 65 82 44
189A, avenue Maurice Schumann | | 59140 Dunkerque, France
===========================================================================
10 years
[OT] how to see if an rpm file will modify config files upon update
by Gianluca Cecchi
Hello,
sorry if this question is more for rpm devs...
I know that in a spec file the definition
%confg ......./file_name1
will cause update of an edited file_name1, saving the on disk one into
.rpmsave before overwriting
while
%confg(noreplace) ......./file_name2
will retain the edited file_name2 on disk and save the provided new config
file into .rpmnew
The question is:
can I see if a config file was defined in its spec with or without
(noreplace) if I have not the spec file and the .src.rpm file, but I have
only the final binary rpm?
I made some tests but I didn't find any difference
Tried queries with "-c" option, with "-V" option and with "--dump" option.
Both either with "-f package.rpm" and directly against rpm db but no
information extracted.
But I think this information should be in any way in the rpm itself...
Thanks in advance
Gianluca
10 years
Remmina again
by Patrick Dupre
Hello,
I am back on my previous message because I do observe another issue.
If I remaisn inactif duringa small while. I ususally cannot
access to the session any more. I just get the login page whitout
been able to give a password.
However, this not not happen 100% of the time.
typically, I have one machine where it almost always happens and one
where it almost never happen.
Now, I have a different session of each.
Thank.
===========================================================================
Patrick DUPRÉ | | email: pdupre(a)gmx.com
Laboratoire de Physico-Chimie de l'Atmosphère | |
Université du Littoral-Côte d'Opale | |
Tel. (33)-(0)3 28 23 76 12 | | Fax: 03 28 65 82 44
189A, avenue Maurice Schumann | | 59140 Dunkerque, France
===========================================================================
10 years