# Emmanuel Touzery's profile - activity

thanks for the tip! turns out dnf can already do that for me with --allowerasing. I'll rather go with that. Too bad the GUI doesn't support that and doesn't hint about that possible option.

2017-07-15 00:48:48 -0500 answered a question upgrade to f26 fails: package conflict

Right so I was using the Software GUI. I had to run it in command-line with the --allowerasing flag as described there: https://fedoramagazine.org/upgrading-...

2017-07-15 00:30:03 -0500 asked a question upgrade to f26 fails: package conflict

i get a package conflict message but the package appears to conflict with itself..? what can I do?

Error running transaction: package libmpeg2-0.5.1-12.fc25.x86_64 is already installed
file /usr/lib64/libmpeg2.so.0.1.0 from install of libmpeg2-0.5.1-12.fc25.x86_64 conflicts with file from libmpeg2-0.5.1-12.fc25.x86_64
file /usr/lib64/libmpeg2convert.so.0.0.0 from install of libmpeg2-0.5.1-12.fc25.x86_64 conflicts with file from libmpeg2-0.5.1-12.fc25.x86_64
file /usr/lib64/libtwolame.so.0.0.0 from install of twolame-libs-0.3.13-5.fc25.x86_64 conflicts with file from twolame-libs-0.3.13-5.fc25.x86_64
file /usr/lib64/libmimic.so.0.0.1 from install of libmimic-1.0.4-9.fc25.x86_64 conflicts with file from libmimic-1.0.4-9.fc25.x86_64
file /usr/share/doc/libmms/ChangeLog from install of libmms-0.6.4-4.fc25.x86_64 conflicts with file from libmms-0.6.4-4.fc25.x86_64
file /usr/lib64/libmms.so.0.0.2 from install of libmms-0.6.4-4.fc25.x86_64 conflicts with file from libmms-0.6.4-4.fc25.x86_64
file /etc/mplayer/mplayer.conf from install of mplayer-common-1.3.0-7.fc25.x86_64 conflicts with file from mplayer-common-1.3.0-7.fc25.x86_64

2016-06-24 13:33:08 -0500 answered a question parrallel install ncurses-6 & ncurses-5 on fedora 24

got the answer in another forum:

use a file search :

# dnf provides */libtinfo.so.5


.... and the reply is ncurses-compat-libs-6.0

Install :

# dnf install ncurses-compat-libs

2016-06-23 16:03:31 -0500 asked a question parrallel install ncurses-6 & ncurses-5 on fedora 24

I just upgraded from fedora 23 to 24, and it broke some binaries I have on my system that were not installed from fedora's repositories, due to the ncurses upgrade from version 5 to version 6:

error while loading shared libraries: libtinfo.so.5: cannot open shared object file: No such file or directory


I wonder whether it's possible to install in parallel ncurses5 & 6 on fedora 24 so I can keep using these binaries? After all that would be the whole point of having these version postfixes, parallel installation right? But I don't see a ncurses5 fedora 24 package...? So.. is it possible?

I started the upgrade from fedora 23 to fedora 24 following this procedure.

But the system-upgrade download fails seems to fail. After I accept the fedora 24 key, the last message is:

Error: Package freetype-freeworld-2.5.5-2.fc24.x86_64.rpm is not signed


Now that's (again..) a rpmfusion package... Anyway, any advice? Wait it out, or disable rpmfusion, or move to unitedrpms, or is there any other option?

2014-08-20 00:49:36 -0500 commented question poor battery life

In fact the laptop has a radeon card besides the builtin intel, however the intel one is used to drive the display and I've made efforts to disable the ATI as much as possible (for battery life, again), and I think I succeeded since the radeon driver doesn't show up in the lsmod.

2014-08-19 10:33:59 -0500 asked a question poor battery life

Find the answer here: battery draining less than one hour

I'm getting a poor battery life out of my dell laptop. like 1h on fedora, 3h on windows. I actually spent a lot energy to reduce power consumption and fan noise and it used to be good at some point, as good as windows I think but now it's poor again.

looking at the powertop, I have 30-35W power consumption. I turned of wired to save power. I have ~8-10W from wireless. 12-15W for the display backlight. Those two are already 20-25W! The next one is at 1.5W... and quickly I have apps.

My display backlight works. I can dim the screen. and it dims itself when idle. I had to force the intel one because at some point it broke, I could not change it anymore. Now with forced intel it works and I can see the power consumption reduces in powertop.

I tried setting power save for the wifi but the performance suffers hugely (like I get hundreds ob kb/s, not anymore mb/s).

Any way to reduce the display backlight power consumption? It's a full HD panel but I feel the number is really bad.

It's a lsmod in case it helps:

Module                  Size  Used by
cpufreq_stats          13344  0
tun                    27153  2
ccm                    17773  2
ip6t_rpfilter          12546  1
ip6t_REJECT            12939  2
xt_conntrack           12760  9
ebtable_nat            12807  0
ebtable_broute         12731  0
bridge                110862  1 ebtable_broute
stp                    12868  1 bridge
llc                    13941  2 stp,bridge
ebtable_filter         12827  0
ebtables               30758  3 ebtable_broute,ebtable_nat,ebtable_filter
bnep                   19624  2
ip6table_nat           13015  1
nf_conntrack_ipv6      18738  6
nf_defrag_ipv6         34712  1 nf_conntrack_ipv6
nf_nat_ipv6            13213  1 ip6table_nat
ip6table_mangle        12700  1
ip6table_security      12710  1
ip6table_raw           12683  1
ip6table_filter        12815  1
ip6_tables             26809  5 ip6table_filter,ip6table_mangle,ip6table_security,ip6table_nat,ip6table_raw
iptable_nat            13011  1
nf_conntrack_ipv4      14656  5
nf_defrag_ipv4         12702  1 nf_conntrack_ipv4
nf_nat_ipv4            13199  1 iptable_nat
nf_nat                 25112  4 nf_nat_ipv4,nf_nat_ipv6,ip6table_nat,iptable_nat
nf_conntrack           99420  8 nf_nat,nf_nat_ipv4,nf_nat_ipv6,xt_conntrack,ip6table_nat,iptable_nat,nf_conntrack_ipv4,nf_conntrack_ipv6
iptable_mangle         12695  1
iptable_security       12705  1
iptable_raw            12678  1
fuse                   91401  5
dell_laptop            18168  0
dcdbas                 14875  1 dell_laptop
arc4                   12608  2
iwldvm                240929  0
mac80211              608109  1 iwldvm
x86_pkg_temp_thermal    14205  0
coretemp               13441  0
kvm_intel             143285  0
kvm                   452397  1 kvm_intel
dell_wmi               12681  0
serio_raw              13434  0
sparse_keymap          13584  1 dell_wmi
videobuf2_vmalloc      13163  0
videobuf2_memops       13161  1 videobuf2_vmalloc
videobuf2_core         43332  0
videodev              147125  1 videobuf2_core
media                  20840  1 videodev
i2c_i801               18145  0
mei_me                 18515  0
iwlwifi               121268  1 iwldvm
crct10dif_pclmul       14307  0
crc32_pclmul           13133  0
crc32c_intel           22094  0
mei                    81536  1 mei_me
ghash_clmulni_intel    13230  0
iTCO_wdt               13480  0
iTCO_vendor_support    13419  1 iTCO_wdt
microcode              44576  0
mii                    13527  0
lpc_ich                21093  0
mfd_core               13182  1 lpc_ich
snd_hda_codec_hdmi     47536  1
snd_hda_codec_conexant    22962  1
snd_hda_codec_generic    67662  1 snd_hda_codec_conexant
bluetooth             429257  5 bnep
snd_hda_intel          30379  4
snd_hda_controller     30139  1 snd_hda_intel
snd_hda_codec         131137  5 snd_hda_codec_hdmi,snd_hda_codec_conexant,snd_hda_codec_generic,snd_hda_intel,snd_hda_controller
snd_hwdep              17650  1 snd_hda_codec
snd_seq                62266  0
snd_seq_device         14136  1 snd_seq
snd_pcm               104333  4 snd_hda_codec_hdmi ...
2014-04-16 04:38:02 -0500 asked a question fedora 20 random freeze

Fedora 20 will freeze randomly for me, it did not happen before. I'm not sure when exactly did it start. When it freezes the HD light is not turned on, so it's not thrashing. ctrl-alt-f1-f7 don't do anything. I can't SSH to the machine from the network.

/var/log/messages tells me nothing useful after a reboot.

i would like to install back the 3.12 kernel to check if it was caused by a recent kernel update but I'm not sure how. Any hint? Also any hint of what else can I do to troubleshoot the problem?

At first I was convinced it only happens when I'm on battery but now it happened once while I was plugged in, for the first time. I'm also considering reinstalling in case something somehow became corrupted. This is a fedora18 installation that I keep upgrading, I'd probably better off with a clean install now.

2014-03-28 02:08:56 -0500 commented answer How to add windows 8 to Grub2

2014-03-03 05:41:49 -0500 answered a question sometimes gdm lists no users

The problem didn't manifest itself since I applied this upgrade: https://lists.fedoraproject.org/pipermail/scm-commits/Week-of-Mon-20140210/1189685.html

So I think that was the cause. I'll remove this answer if the problem occurs again, therefore proving me wrong.

2014-02-17 14:36:50 -0500 commented question sometimes gdm lists no users

thanks. my user id is 1000. it's the typical case I guess, single user on this laptop. it's reproducible, I reboot a couple of times and I get both cases I think. I would expect there to be some log somewhere... I tried but I only find the X11 logs. I'm trying to reinstall now, I'll see if it makes any difference, thanks for the tip!

2014-02-17 13:38:00 -0500 asked a question sometimes gdm lists no users

I'm using fedora 20. Since a few upgrades ago, sometimes after boot gdm will not display any users and so log in is impossible. If I go in a text terminal with ctrl-alt-f2 and log in, then run sudo killall gdm, gdm restarts and does show my single user so that i can login.

I have opened this bug related to this, but no reaction so far: https://bugzilla.redhat.com/show_bug.cgi?id=1065687

EDIT: sadly yum reinstall didn't help. I hoped a bit after the first reboot helped, but now the problem happened again. Also, this is just a plain user as created by the fedora installer, local user, nothing special, all standard.

And here is the output of sudo journalctl -u gdm (otherwise the log is superlong, I'm using this installation for a long time, this is a F18 upgraded to F19 upgraded to F20...).

I think this is a boot where everything went OK:

-- Reboot --
Feb 17 21:50:00 emmanuel-linux systemd[1]: Starting GNOME Display Manager...
Feb 17 21:50:01 emmanuel-linux systemd[1]: Started GNOME Display Manager.
Feb 17 21:50:07 emmanuel-linux gdm[666]: Failed to give slave programs access to the display. Trying to proceed.
Feb 17 21:50:36 emmanuel-linux systemd[1]: Stopping GNOME Display Manager...
Feb 17 21:50:37 emmanuel-linux systemd[1]: Stopped GNOME Display Manager.


And in this case there was a problem:

-- Reboot --
Feb 19 07:49:39 emmanuel-linux systemd[1]: Starting GNOME Display Manager...
Feb 19 07:49:40 emmanuel-linux systemd[1]: Started GNOME Display Manager.
Feb 19 07:49:47 emmanuel-linux gdm[675]: Failed to give slave programs access to the display. Trying to proceed.
Feb 19 07:50:21 emmanuel-linux systemd[1]: gdm.service holdoff time over, scheduling restart.
Feb 19 07:50:21 emmanuel-linux systemd[1]: Stopping GNOME Display Manager...
Feb 19 07:50:21 emmanuel-linux systemd[1]: Starting GNOME Display Manager...
Feb 19 07:50:21 emmanuel-linux systemd[1]: Started GNOME Display Manager.
Feb 19 07:50:21 emmanuel-linux gdm[1741]: Failed to give slave programs access to the display. Trying to proceed.
Feb 19 07:50:34 emmanuel-linux gdm[1741]: Failed to remove slave program access to the display. Trying to proceed.


But probably the difference is only because I restarted gdm with the killall.

In var/log/gdm, the only files which have been recently updated are :0.log and :0.log.1 and this is the X11 log, I see nothing interesting there.

_EDIT:_ well i'm currently installing the f20 upgrade which has this fix inside: https://lists.fedoraproject.org/pipermail/scm-commits/Week-of-Mon-20140210/1189685.html

So I'm kind of hoping the issue could be related and fixed by this problem. I have those hangs currently in maybe 20% of the boots.

2013-12-21 13:30:36 -0500 asked a question fedora 20: cups not shutting down

since I upgraded to fedora 20, my machine is not shutting down. by pressing ESC at the plymouth screen for shutting down, I saw it's the cups service which does not stop. seems to think it's printing something maybe.

once i rebooted, i see that cups is not stopping properly with systemctl status I can see that the stopping times out, then systemd kills it, but only after the timeout which is 1:30. Annoying to have to wait 1:30 for no reason everytime I shut down the computer.

I checked and I think there are no jobs left in the queue that CUPS may be waiting on. "lpstat -t" is telling me both the printers i have configured are idle. Actually i have only one USB printer, a HP printer, but I have it configured both as usb:// and hp:/ I usually print on the usb:// but the hp:/ one allows me to see ink levels and so on with the HP device manager.

Any hint on what I could do to solve this problem?

2013-07-05 13:05:08 -0500 edited answer How to start a systemd service, after DHCP has initialized the IP-Adress?

Well, on fedora 19 the other solutions didn't work for me (well I didn't try the one that delays all services just for that... I only want to delay minidlna because it's the only problematic service), so i gave up and went for the ugly-but-it-works solution.

I wrote this script:

#!/usr/bin/sh
ip_row=ifconfig | grep 192
while [ ! -n "\$ip_row" ]
do
ip_row=ifconfig | grep 192
sleep 2
done

# OK now there an IP...
/usr/sbin/minidlna -f /etc/minidlna.conf -P /var/run/minidlna/minidlna.pid


I know my IP is going to be 192.*, so I relied on that. Once ifconfig returns at least one row containing "192" i know I have an IP. So I make a loop and once I have an IP then I launch minidlna. Works for me.

So then in /usr/lib/systemd/system/minidlna.service I changed the ExecStart row to start this script instead of starting minidlna directly.

Remember to give exe rights to that script (chmod 755), and you can debug any problems by checking in /var/log/messages and by running the very helpful systemctl status command.