From: Greg Woods [mailto:email@example.com]
Sent: zaterdag 25 februari 2017 20:45
To: Community support for Fedora users
Subject: Re: automatically restart ethernet ?
On Sat, Feb 25, 2017 at 11:39 AM, fred roller <fredroller66(a)gmail.com<mailto:firstname.lastname@example.org>> wrote:
Look into "heartbeat".
"heartbeat" has been deprecated for several years now and is no longer supported. There are no longer heartbeat packages in the F25 repos. The supported way to do HA on Linux is now corosync with pacemaker. But this is a pretty big thing to swallow just for restarting a net connection. You might want to try a script that you can run more often than hourly that first checks to see if the net is up (maybe with a ping to your default gateway?), and only restarts NetworkManager if it's not. There are also other monitoring applications such as "monit", but that's a pretty complex beast as well.
I presume you've looked into why your server keeps losing the net connection? That could be a bad cable, a bad switch port, or a bad NIC. Hardware issues tend to get worse over time; if that's what it is, even a cron kludge may eventually not work any more.
Pinging a local machine should tell you when machines become unreachable, probably because of a network failure.
But, as said before, it only treat consequences, not the real cause. Perhaps you should focus on that.
Why does restarting NM help? To bring the interface up again? Did it loose its IP-address for some reason? Duplications?
Did you consider / try a fixed IP-address?
Dit bericht kan informatie bevatten die niet voor u is bestemd. Indien u niet de geadresseerde bent of dit bericht abusievelijk aan u is toegezonden, wordt u verzocht dat aan de afzender te melden en het bericht te verwijderen. De Staat aanvaardt geen aansprakelijkheid voor schade, van welke aard ook, die verband houdt met risico's verbonden aan het elektronisch verzenden van berichten.
This message may contain information that is not intended for you. If you are not the addressee or if this message was sent to you by mistake, you are requested to inform the sender and delete the message. The State accepts no liability for damage of any kind resulting from the risks inherent in the electronic transmission of messages.
I have a new USB3 drive and would like to install Windows on one
partition of it for occasional dual booting. I already have F25 on my
main drives (root on a 128GB SSD and /home on a 1TB internal SATA) so
the usual advice of "install Windows first, then Linux" is not really
what I want to hear unless there's no alternative. My mobo can support
BIOS or UEFI.
Any pointers to how to go about this would be appreciated, (including
getting Windoze to run from a USB drive, and apologies for being OT
with that part).
I am trying to boot Fedora 25 live media on a MITAC PD11TI board. By
removing "quiet" from the kernel command line, I am able to see the
kernel messages up until KMS kicks in. After that, all I get is a black
screen with a "no signal" message.
I've tried adding both "nomodeset" and "3" to the kernel command line,
and I've tried booting with "basic graphics," all to no effect.
How the bleep can I boot Fedora 25 live media into text mode with no
KMS? (This is an Atom N2800 with an integrated PowerVR GPU, with seems
to use the gma500_gfx module; this module does not appear to have any
Ian Pilcher arequipeno(a)gmail.com
-------- "I grew up before Mark Zuckerberg invented friendship" --------
Would it be fair to characterize the state of F25 development for ARM as
"experimental". It's certainly not ready for prime time.
I installed the MATE ARM spin to a 128G Class 10 microSD rated for 85MB/s
reads (not rated for writes). I was delighted to see my RP3 boot up
perfectly to the Anaconda config screens. But after finishing the
installation I very soon started to see the desktop freeze up solid. The
only way to recover is to cycle power, and that frequently requires me to
use the Recovery boot.
Any time I return to the RP3 after about ten idle minutes, I find it is
locked up. This is obviously not what I had in mind when I bought it. Is
this a known problem, one with a work-around?
Fairview Heights, IL
On F25, a server keeps losing its internet connection. I've now added a
cron.hourly script to restart NetworkManager. Is there a better way ?
Some way to tell NetworkManager that if the connection goes down, it
should try to bring it up every ___ seconds ?
I'm have a problem where when I am using my logitech MX Anywhere 2 mouse
with fedora using bluetooth, the mouse pairs, connects and works fine for
around a minute or 2 and then stops working.
I have the following output in dmesg:
[220343.835317] input: MX Anywhere 2 as
[220343.840166] hid-generic 0005:0000:0000.0035: input,hidraw1: BLUETOOTH
HID v0.00 Keyboard [MX Anywhere 2] on F0:D5:BF:EC:71:19
[220443.987024] Bluetooth: hci0 command 0x0c1a tx timeout
[220446.034387] Bluetooth: hci0 command 0x200c tx timeout
Some internet searching leads me to think this could be related to some
sleep functionality in BTLE. How could I further diagnose/resolve the
issue? I need to turn bluetooth off and on to regain functionality (which
once again only lasts a minute). As far as hardware is concerned, its some
sort of intel integrated wireless chipset in a thinkpad:
# lsmod | grep -i blue
bluetooth 528384 33 btrtl,btintel,bnep,btbcm,rfcomm,btusb
rfkill 24576 8 bluetooth,thinkpad_acpi,cfg80211
I just switched my son's computer to one that has a Radeon 5450. A low-end
card, but still it should be better than his old computer's Intel GMA 2500.
It's running the radeon open source driver and performance in minecraft
(the only app that's relevant) is worse than before.
I couldn't just take the video card and put it in his old computer--it's a
small form-factor with no slots.
This is Fedora 25, like it was before. The new computer has the same
processor as the old, though less RAM (5GB v. 8GB--unfortunately I was not
able to swap it over).
I see lots of conflicting info online about whether the fglrx driver works
on newer kernels, and I don't want to do anything time-consuming to get it
to work. Is it easier than I think? Should I just buy a cheap NVIDIA card,
or will that be a hassle too? Or is the RAM the issue?
All advice welcome. Thanks!
Had a system crash, and was wondering, in the fed/centos world (v6 and
above -guess that would be fed 22??) is there a process that can
basically scan the system, so that if the system crashes, everything
would be auto started on restart.
Also, would it allow the gedit processes to be fired up with the
requisite files being edited??
I saw something for ubuntu - dconf-editor, and I've seen some things
within the gconf where you can specify certain processes to be run on
But I'm more looking for something I can run, so that if I have a
crash, the system is pretty much back running prior to the crash.
# uname -a
Linux machine 4.9.9-200.fc25.x86_64 #1 SMP Thu Feb 9 17:28:13 UTC 2017
x86_64 x86_64 x86_64 GNU/Linux
# man dnf.plugin.reposync
No manual entry for dnf.plugin.reposync
Copying arguments from my working (YUM) reposync command doesn't work.
# dnf reposync -r google-chrome --norepopath -m
usage: dnf reposync [--help-cmd] [-p DOWNLOAD_PATH] [--repo REPO]
--help-cmd show this help about this tool
-p DOWNLOAD_PATH, --download-path DOWNLOAD_PATH
where to store downloaded repositories
--repo REPO repository to download
Error: unrecognized arguments: -r google-chrome --norepopath -m
Trying a simple command per the help output above.
# dnf reposync --repo google-chrome
usage: dnf [--allowerasing] [-b] [-C] [-c [config file]] [-d [debug level]]
[--debugsolver] [--showduplicates] [-e ERRORLEVEL]
[--rpmverbosity [debug level name]] [-q] [-v] [-y] [--assumeno]
[--version] [--installroot [path]] [--enablerepo [repo]]
[--disablerepo [repo]] [-x [package]] [--disableexcludes [repo]]
[--repofrompath [repo,path]] [--noplugins] [--nogpgcheck]
[--disableplugin [plugin]] [--color COLOR]
[--releasever RELEASEVER] [--setopt SETOPTS] [--refresh]
Command line error: argument --repofrompath: bad format: google-chrome
Evidently the command syntax is a bit confused. I get different ideas
from the two commands.
Can anyone point me to the real documentation of this command? Or tell
me how to translate my YUM command to the DNF equivalent?