where do dnf plugins go?
by Tom Horsley
The DNF documentation desperately attempts to avoid actually
saying anything :-(. For pluginpath it says:
List of directories that are searched for plugins to load. Plugins found in any of the directories in this configuration option are used. The default contains a Python version-specific path.
The installed /etc/dnf/dnf.conf file doesn't explicitly say anything
about pluginpath, so it must be hard coded somewhere.
So what the heck is the actual contents of this pluginpath config
setting on fedora 22? Where do I install a .py module if I'm
trying to write my own plugin?
8 years, 10 months
Building/loading dahdi from asterisk
by Alex
Hi,
I'm really stuck with getting asterisk started. I've built and install
the dahdi asterisk modules, but they don't load automatically on boot.
The asterisk-dahdi package includes /etc/rc.d/init.d/dahdi which runs
"/bin/systemctl start dahdi.service" but it fails to load the modules.
My question is, how do I troubleshoot this? How do I get more
information to find out why it isn't automatically load the modules?
If I just run "modprobe dahdi" manually, it loads the modules.
I've looked through the logs, and it just says "Started LSB: DAHDI
kernel modules." without any further information.
Any ideas greatly appreciated.
Thanks,
Alex
8 years, 10 months
Fedora-22 KDE panel
by Timothy Murphy
I've just installed Fedora-22 from a Fedora-22 Live KDE usb stick.
After dnf-installing a few packages, most things seem to be working,
but I'm puzzled by the apparent lack of a panel,
or if the bar at the top with Activities and Time are the panel
then I'm puzzled by the lack of icons in the panel.
I read advice to right-click in the panel,
but if this is the panel that has no effect.
Also I miss the multiple desktops I had in Fedora-21 KDE.
Have these been abolished in Plasma-5?
A very ignorant question - I see everyone refers to KDE Plasma.
Is there in fact another non-Plasma version of KDE
that one could run under Fedora?
--
Timothy Murphy
gayleard /at/ eircom.net
School of Mathematics, Trinity College, Dublin
8 years, 10 months
Silencing auditd in fedora22
by Alex
Hi,
Since upgrading from fedora22, auditd is drowning /var/log/messages
with useless information such as this:
Jul 18 19:02:19 orion audit: <audit-2404> pid=6002 uid=0
auid=4294967295 ses=4294967295 msg='op=destroy kind=server
fp=SHA256:b5:7b:76:df:38:16:f3:f5:cd:2f:67:54:9a:2e:68:15:ae:9c:40:50:4f:6d:81:43:0d:54:bd:e2:c5:a0:43:7f
direction=? spid=6002 suid=0 exe="/usr/sbin/sshd" hostname=?
addr=64.1.XX.18 terminal=? res=success'
I've enabled rsyslog because the logs are so much easier to access,
but I'm not using auditd so would like to just turn it off.
Ideas for using journalctl to show me the following would be appreciated:
- start at the end of the log
- use shortened hostname
- shows only info in the former /var/log/maillog or /var/log/messages
- piped through a searchable pager
Typing "less /var/log/maillog" requires far less typing, more easily
remembered, and is easily searchable.
Thanks,
Alex
8 years, 10 months
w/out clean all, dnf does nto find packages to be updated
by Maurizio Marini
Here is F22
Maybe this was already posted on list.
I have plugin that notify some packges should be updated:
Software Updates
You have 19 new updates
using dnf update
or dns upgrade
it does not find anything:
--------------------
# dnf update
Last metadata expiration check performed 1:27:33 ago on Sun Jul 19 19:54:29
2015. Dependencies resolved.
Nothing to do.
Complete!
--------------------
then I do a clean all
--------------------
# dnf clean all
Cleaning repos: fedora updates
Cleaning up Everything
--------------------
Now dns find 19 packages to be updated:
---------------------------
# dnf update
Fedora 22 -
x86_64 300 kB/s | 41 MB 02:20
Fedora 22 - x86_64 -
Updates 530 kB/s | 12 MB 00:23
Last metadata expiration check performed 0:00:25 ago on Sun Jul 19 21:28:39
2015.
Dependencies resolved.
[....]
Upgrade 19 Packages
Total download size: 2.9 M
Is this ok [y/N]:
-----------------------------
-m
8 years, 10 months
Problem pluggin smartphone
by Sergio Belkin
Hi Dear community,
I have a problem when pluggin smartphone, as journalctl shows below:
jul 19 17:28:43 newhope.belkin.home kernel: usb 1-1: new full-speed USB
device number 2 using xhci_hcd
jul 19 17:28:43 newhope.belkin.home kernel: usb 1-1: device descriptor
read/64, error -71
jul 19 17:28:43 newhope.belkin.home kernel: usb 1-1: device descriptor
read/64, error -71
jul 19 17:28:44 newhope.belkin.home kernel: usb 1-1: new full-speed USB
device number 3 using xhci_hcd
jul 19 17:28:44 newhope.belkin.home kernel: usb 1-1: device descriptor
read/64, error -71
jul 19 17:28:44 newhope.belkin.home kernel: usb 1-1: device descriptor
read/64, error -71
jul 19 17:28:44 newhope.belkin.home kernel: usb 1-1: new full-speed USB
device number 4 using xhci_hcd
jul 19 17:28:44 newhope.belkin.home kernel: usb 1-1: Device not responding
to setup address.
jul 19 17:28:44 newhope.belkin.home kernel: usb 1-1: Device not responding
to setup address.
jul 19 17:28:45 newhope.belkin.home kernel: usb 1-1: device not accepting
address 4, error -71
jul 19 17:28:45 newhope.belkin.home kernel: usb 1-1: new full-speed USB
device number 5 using xhci_hcd
jul 19 17:28:45 newhope.belkin.home kernel: usb 1-1: Device not responding
to setup address.
jul 19 17:28:45 newhope.belkin.home kernel: usb 1-1: Device not responding
to setup address.
jul 19 17:28:45 newhope.belkin.home kernel: usb 1-1: device not accepting
address 5, error -71
jul 19 17:28:45 newhope.belkin.home kernel: usb usb1-port1: unable to
enumerate USB device
lsusb output is as follows:
[sergio@newhope ~]$ lsusb
Bus 004 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub
Bus 004 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 005: ID 1bcf:289a Sunplus Innovation Technology Inc.
Bus 003 Device 004: ID 0cf3:e004 Atheros Communications, Inc.
Bus 003 Device 003: ID 0bda:0129 Realtek Semiconductor Corp. RTS5129 Card
Reader Controller
Bus 003 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Also, youn can take a look to the usb-devices output at
http://pastebin.com/NHD5WBSu
Please could you help me?
Using Fedora 22: kernel 4.0.8-300.fc22.x86_64
Thanks in advance!
--
--
Sergio Belkin
LPIC-2 Certified - http://www.lpi.org
8 years, 10 months
Systemd user services
by Suvayu Ali
Hi,
I wanted to run systemd user services, in linger mode, meaning they run
as a regular user and hang around without an active session. This
worked nicely for the default target. When I changed to a different
target however, the systemd user is not working anymore. Any thoughts?
This is what I did step by step:
- Enable linger mode and reboot:
# loginctl enable-linger user
- Now from a console login when I do:
$ systemctl --user status
I see systemd user service is running.
- I add a service like this:
$ systemctl --user enable emacs
$ systemctl --user start emacs
Looking at the status now, I see emacs is running as expected.
This was just a test, I actually want to run mpd, but I was not getting
any sound without a graphical session, so temporarily I switched to a
graphical target with:
# systemctl isolate graphical.target # my default is runlevel3
IIUC, when I do this all services get restarted. So I wasn't surprised
when there was not running emacs (checked with pgrep). I was surprised
however to see I cannot query the user session anymore.
$ systemctl --user enable emacs.service
Failed to get D-Bus connection: No such file or directory
It is even more surprising since the systemd user service is still running!
$ ps -uf -p 14301
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
lightdm 14301 0.0 0.1 45004 4864 ? Ss 10:18 0:00 /usr/lib/systemd/systemd --user
What is wrong? Am I mistaken somewhere, or is this a bug?
Thanks for any thoughts.
--
Suvayu
Open source is the future. It sets us free.
8 years, 10 months
"dnf update" just now seems to have broken dnf
by Robert P. J. Day
"dnf update" just now updated, among other things,
dnf-langpacks.noarch 0.11.0-1.fc22, at which point running "dnf
update" again produces:
# dnf update
Traceback (most recent call last):
File "/bin/dnf", line 36, in <module>
main.user_main(sys.argv[1:], exit_code=True)
File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 185, in user_main
errcode = main(args)
File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 84, in main
return _main(base, args)
File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 115, in _main
cli.configure(map(ucd, args))
File "/usr/lib/python2.7/site-packages/dnf/cli/cli.py", line 984, in configure
self.base.plugins.run_init(self.base, self)
File "/usr/lib/python2.7/site-packages/dnf/plugin.py", line 104, in run_init
plugin = p_cls(base, cli)
File "/usr/lib/python2.7/site-packages/dnf-plugins/langpacks.py", line 692, in __init__
conflist = config.get('main', 'langpack_locales')
File "/usr/lib/python2.7/site-packages/iniparse/compat.py", line 222, in get
value = RawConfigParser.get(self, section, option, vars)
File "/usr/lib/python2.7/site-packages/iniparse/compat.py", line 125, in get
raise NoOptionError(option, section)
ConfigParser.NoOptionError: No option u'langpack_locales' in section: u'main'
#
*that* doesn't look good ...
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
8 years, 10 months
Excessive verbosity
by David A. De Graaf
The verbosity of systemd and its associated logging functions is out
of control. It drives me crazy, and I can only hope that this attempt
to complain constructively will be heeded.
After recovering from the disaster of filling my root filesystem while
away from home, I'm attempting to bring back some discipline to
/var/log. Aside from the outrage of the journal system,
/var/log/httpd/error_log and /var/log/audit/audit.log are the worst
offenders.
The httpd errors are of this recurring general form:
[Sat Jul 18 10:48:39.918368 2015] [:error] [pid 10957] [client
192.168.2.1:54848] PHP Deprecated: Non-static method
GalleryUtilities::isEmbedded() should not be called statically,
assuming $this from incompatible context in
/usr/share/gallery2/modules/core/classes/GalleryUrlGenerator.class
on line 381, referer: http://datix.us/
and seem to be due to a major screwup in the gallery2 code with
respect to static vs. non-static methods, whatever that means.
A whack-a-mole debugging process is ongoing to fix these.
Thanks to Joel for responding to BZ 1214128 and trying to fix 'em.
The auditd system seems grotesque to me. It purports to insert code
in every single system call and report on it. This can only seriously
degrade Linux's efficiency. Others have noticed the excessive volume of
reporting and tried to suppress it. One recommends removing the audit
package, or disabling auditd, but that merely transfers the reporting
to /var/log/messages.
In https://fedorahosted.org/fesco/ticket/1311 I read that the overhead
is so great that it was
"AGREED: turn off syscall auditing for all processes by default."
with this to be done by
Add something like this to /etc/audit/rules.d/audit.rules:
# This suppresses syscall auditing for all tasks started
# with this rule in effect. Remove it if you need syscall
# auditing.
-a task,never
Sure enough, that's the way Fedora 22 is delivered.
However, it doesn't work!
The audit.log is still swamped with unwanted drivel.
Following a clue in
https://www.redhat.com/archives/linux-audit/2012-January/msg00012.html
I have added these commands to /etc/rc.d/rc.local with good results:
# Stop auditd from writing so much crap
auditctl -a exclude,always -F msgtype=MAC_IPSEC_EVENT
auditctl -a exclude,always -F msgtype=USER_AUTH
auditctl -a exclude,always -F msgtype=USER_ACCT
auditctl -a exclude,always -F msgtype=CRED_REFR
auditctl -a exclude,always -F msgtype=CRED_DISP
auditctl -a exclude,always -F msgtype=CRED_ACQ
auditctl -a exclude,always -F msgtype=USER_START
auditctl -a exclude,always -F msgtype=USER_END
auditctl -a exclude,always -F msgtype=SERVICE_START
Now the only msgtype left in audit.log is type=LOGIN, which occurs
whenever a cronjob fires off. Since I'm a big user of cronjobs, there
are still too many of these.
In whose perverted mind is running a cronjob related in any way to
logging in?
While venting, there are two other problems with the audit system:
1) The auditctl man page says to combine multiple -F options in a
single command for efficiency. It doesn't work.
2) With all the verbosity of audit.log, nowhere is the date shown.
If this data were really intended to be used or read, wouldn't it be
possible to squeeze in `date` somehow?
--
David A. De Graaf DATIX, Inc. Hendersonville, NC
dad(a)datix.us www.datix.us
"Those who hear not the music, think the dancers mad."
8 years, 10 months
dnf seems to be broken after the most recent updates, including updates-testing
by Joachim Backes
Hi all F22 users,
it seems dnf is broken after the most recent updates (including
updates-testing):
-------------------------------------------------------------------------
sudo dnf update
Traceback (most recent call last):
File "/bin/dnf", line 36, in <module>
main.user_main(sys.argv[1:], exit_code=True)
File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 185, in
user_main
errcode = main(args)
File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 84, in main
return _main(base, args)
File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 115, in
_main
cli.configure(map(ucd, args))
File "/usr/lib/python2.7/site-packages/dnf/cli/cli.py", line 984, in
configure
self.base.plugins.run_init(self.base, self)
File "/usr/lib/python2.7/site-packages/dnf/plugin.py", line 104, in
run_init
plugin = p_cls(base, cli)
File "/usr/lib/python2.7/site-packages/dnf-plugins/langpacks.py", line
692, in __init__
conflist = config.get('main', 'langpack_locales')
File "/usr/lib/python2.7/site-packages/iniparse/compat.py", line 222,
in get
value = RawConfigParser.get(self, section, option, vars)
File "/usr/lib/python2.7/site-packages/iniparse/compat.py", line 125,
in get
raise NoOptionError(option, section)
ConfigParser.NoOptionError: No option u'langpack_locales' in section:
u'main'
---------------------------------------------------------------------------
yum-deprecated still works!
Anybody sees this too?
Kind regards
Joachim Backes
--
Fedora release 22 (Twenty Two)
Kernel-4.1.2-200.fc22.x86_64
Joachim Backes <joachim.backes(a)rhrk.uni-kl.de>
https://www-user.rhrk.uni-kl.de/~backes
8 years, 10 months
