tl;dr story is that the system firmware has some variables that say where
to look for EFI volumes, and it launches EFI executables from those volumes. No
tiny boot sectors or second stage loaders, just EFI executables on a FAT32
filesystem.
--
Jonathan Billings
--
__
last year?
Try disabling the timeout with:
echo LIBVIRTD_ARGS= > /etc/sysconfig/libvirtd && systemctl restart libvirtd
--
Jonathan Billings--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@list
who were involved with the discovery and investigation. We should also all
thank Andres Freund for his meticulous discovery of the backdoor, without
which, we might have ended up with it he backdoor running in production for
many distros.
--
Jonathan Billings
--
_
ootloaders or partition tables, although it
probably doesn’t hurt. I had a kickstart that preserved custom stuff like the
krb5 keytab between reloads in the kickstart %pre section, so I didn’t want to
just nuke the filesystem.
--
Jonathan Billings
--
_
com/ReimuNotMoe/ydotool) but they
need direct access to /dev/input, which generally need root access (and if
every app can read it, it’s just as bad as X).
I suspect this is something a GNOME extension would replace, and not by sending
key presses but replicating the results.
--
Jonathan Billin
” in some form. Perhaps there was an automatic translation error?There was a “castor” package once but it has been abandoned for at least 3 years. (https://src.fedoraproject.org/rpms/castor/commits/rawhide )-- Jonathan Billings--
___
users mailing list
s the alpha or beta release and is always
subscribed to the testing repos.
Honestly, when I first looked at the script I figured it was the wget->wget2
update that landed in F40.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fed
s the
saved hostname.
I’ve never encountered one of these devices, but I’ve heard of windows systems
and even phones can do this.
Set your preferred hostname either in GNOME Settings (or whatever DE’s
settings) or with hostnamectl.
--
Jonathan Billings
--
_
te results, so I imagine they’ll always have ways to
track users and post numbers.
--
Jonathan Billings--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora
e
> the problem then, or how to further troubleshoot?
>
By “SFTP” do you mean FTP over SSL/TLS, or the OpenSSH “sftp” backend which
runs over the SSH protocol? Two completely different protocols.
I suspect that lftp just calls out to the OpenSSH sftp binary i
ing over the SSH port and all access is handled via the
SSH’s config.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https
ase describe the
actual errors you get, and include the appropriate errors you find in the httpd
error_log.
--
Jonathan Billings--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedo
probably use a kickstart to automate an install, but that’s what I do for
my job so it’s my comfort zone.
--
Jonathan Billings
> On May 31, 2024, at 21:06, Alex wrote:
>
>
> Hi,
> I have a new server with OVH that I'd like to install fedora40 over the
> network in some
a separate EFI boot entry for each
part of the array would get you the closest to redundancy. But it would still
only be looking at the fat32 filesystem on each device.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To
If so, then you can manually run a relabel by running “sudo
restorecon -R -v /etc”.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Condu
On Jun 19, 2024, at 16:41, Joe Zeff wrote:
>
> On 06/19/2024 02:31 PM, Jonathan Billings wrote:
>> So, if you’ve used a rescue disk to boot into Linux, chroot into your OS’s
>> disk, and run “passwd” to change a password, or otherwise edit /etc/shadow
>> or /etc/passwd
e if it's something inhibiting a shutdown, but that's how you'd check.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
move it from the directory to get it to
stop executing a second cron job.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https
; {
> chainloader (hd1)+1
> }
>
> but changed to this:
>
> menuentry "Boot from first harddisk" {
> chainloader (hd0)+1
> }
>
> Give it a try and see what happens.
That won’t work for UEFI, you’d need to give a path to a UEFI executable on
hdX.
I believe
On Jul 3, 2024, at 12:15, Alex wrote:
> However, I also have virtual interfaces on eth1:1 and others. What are the
> next steps?
Interface aliases are pretty much deprecated (see the thread about ifconfig),
you’d just assign multiple IPs/prefixes to eth1.
--
Jonathan Bi
a non-efi bootloader if you booted via EFI bootloader on installer.
That would explain why there’s no EFI volume on your 36 install.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to user
ted the bootloader.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
L
written executable code that would allow remote login (such as dropping a pubkey into a user’s authorized_keys) or by altering an admin’s login environment (replacing their .bashrc). Running web sites out of /home is still possible, they just can’t also write to /home.-- Jonathan
w to get out of this trap?
>
> Thank you.
I think it also wants a /boot partition in this case. It can be ext4. Make
sure it is big enough for 3 kernels+initrds, at least 1G.
I’ve not really explored the RAID install options so it might be that the type
of RAID you chose can’t be mo
ot supported in
the default configuration. Is it possible that whoever issued the cert has
created a new one with an algorithm that isn’t rather old and falling out of
support?
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedorapr
re interested in something like OpenSCAP security profiles, which
Fedora suppets:
https://static.open-scap.org/ssg-guides/ssg-fedora-guide-standard.html
--
Jonathan Billings--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe
some how-to explaining with details how to proceed. I'm not very
> good with efi boot system nor secure boot.
You have to enroll your own key in the firmware:
https://rpmfusion.org/Howto/Secure%20Boot
--
Jonathan Billings--
> I wonder if anyone vets that information?
Just check the references:
https://static.open-scap.org/ssg-guides/ssg-fedora-guide-standard.html#xccdf_org.ssgproject.content_rule_accounts_maximum_age_login_defs
--
Jonathan
--
___
users mailing list -- u
On Jul 15, 2024, at 07:35, Neal Becker wrote:
>
>
> I've noticed that recently I'm not seeing Discover notifying me of available
> updates. Is there some setting for this?
Does ‘sudo dnf update’ return any updates either?
What release of Fedora are you using?
RHEL, perhaps you can open a case with Red Hat to get help?
There is also a 389-users list, although I don’t know if it’s limited to Fedora
versions.
https://lists.fedoraproject.org/archives/list/389-us...@lists.fedoraproject.org/
--
Jonathan Billings--
___
package but not included, this is probably not the right label for this kind of
file.
Rkhunter complains about anything it finds in /dev/shm.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an
s, but I'm not sure if I need to run grub2-
> mkconfig as well.
The “btrfs-replace” command can be used to do this. I use it in automation to
migrate a btrfs filesystem on an unencrypted volume to a luks volume. All on
one disk.
You do need to manually recreate the efi and boot partit
is just one of the many entries in the Desktop file.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraprojec
install the sdubby package and re-run
the kernel-install command, or more simply, reinstall the kernel-core package.
Once that's done, it will do the normal GRUB2 boot method of putting files into
/boot/ and blscfg entries in /boot/loader/entries/.
--
Jonathan Billings
--
on the EFI volume or preventing any future attempts at booting from those
files.
--
Jonathan Billings
--
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Co
diskinfo/files/"; -o - |
> grep -i '9.4.5'
Use “grep -F”
-F, --fixed-strings
Interpret PATTERNS as fixed strings, not regular
expressions.
--
Jonathan Billings--
___
users mailing list -- users@list
s too nice outside here to get
out of my hammock!)
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproject.org/
does a lot of the hard work of figuring that out
for you, so you don't need to manually enter that information.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users
0003 boot entry. The order
is already defined.
The UEFI VBOX HARDDISK entries are most likely the VBox UEFI
implementation's entries to boot via the Compatibility Support Module
(CSM), which is the legacy BIOS boot method.
--
Jonathan Billings
___
g that the
systemd-fstab-generator parses when it dynamically creates .mount
units from fstab entries. For what it's worth, it's probably a better
idea to just create them as .mount units, that way you can take
advantage of them as requirements or dependencies for other
stall the bootloader
that was used to boot the install medium.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fed
g as root, can't poke around
in the gvfs mounts for a user. The FUSE mounts for gvfs is locked
down so only users can get at them, as a security measure. lsof looks
at all processes, including the gvfs ones.
--
Jonathan Billings
___
users mailing
On Wed, Jun 03, 2020 at 11:37:51AM -0500, Roger Heflin wrote:
>
> It is not much of a useful security measure, since root can su - user
> and take a look see anyway.
It's more for protecting against root-level services and not a
malicious admin running as root.
--
Jon
lso were
dropped, which is why it isn't able to be installed anymore.
1. https://bugzilla.redhat.com/show_bug.cgi?id=1736646
2. https://bugzilla.redhat.com/show_bug.cgi?id=1834103
--
Jonathan Billings
___
users mailing list
starting a root shell with sudo? Or were you using
root at the console login instead of the graphical login?
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject
people who still do it, and they get
compromised that much easier.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedora
options.
Maybe this is a bug that has been fixed since I encountered it?
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://do
;ve
booted via EFI. If it doesn't? Then you aren't.
You might also want to make sure there aren't any other EFI or /boot
partitions that might be hanging around that are being used by the
boot that have old data on them.
--
Jonathan Billings
_
On Wed, Jun 10, 2020 at 12:20:35PM -0600, linux guy wrote:
> On Wed, Jun 10, 2020 at 7:31 AM Jonathan Billings
> wrote:
>
> > First, it's probably worth seeing whether you're booting via a legacy
> > boot (CSM) or directly to UEFI.
> >
> > Does /s
ase dump and restore described there might help you.
I'd be interested in seeing the output of:
echo 'PRAGMA integrity_check;' | sudo sqlite3 /var/lib/dnf/history.sqlite
It should just return:
ok
--
Jonathan Billings
___
use
On Jun 19, 2020, at 03:34, Stephen Morris wrote:
>
> My issue with BLS is the way it displays the kernel menu entries in grub, it
> is exactly the same as the way grubby showed the entries which I also hated,
> consequently I refuse to use BLS until such time as they provide an option to
> ge
different shell or the permissions only allows root.
The output of “rpm -V bash” should tell you what is wrong with the package.
Now, personally, I see a long out of date system with ssh access and broken
bash, and I immediately think “compromised system”. So I’d suggest caution.
--
Jonathan Bil
ter:
$ printf '%02x-%02x-%02x\n' 6 94 3
06-5e-03
I do see it. And the microcode_ctl update does appear to be applying
an update. Actually, that one is a RHEL system and in its notes, it
says that it doesn't apply the microcode update because it cause
On Wed, Jul 01, 2020 at 07:10:15PM +0100, Simon Colston wrote:
>
> I guess nobody has experience of this. Is this not normal practice?
I've been using autofs with Kerberos authentication coming from the
user's tickets (and the request-key infrastructure for credential
handof
ate build target. I didn't get too much more into
it, but I think its safe to say, you should report flatpak issues on
the audacity component. Hopefully the person in charge of it is even
aware that it's being built as a flatpak.
--
Jonathan Billings
__
On Jul 6, 2020, at 00:33, ToddAndMargo via users
wrote:
> my virt-manager starts as a user now, but the VM's
> I configured are gone (back as root)
When you run virt-manager as root, which connection are you using? Is it
“qemu:///system” or “qemu:///session”?
--
Jonath
thenaeum uses flatpak as its packaging system and pulls all data from
> flathub currently.
GNOME software updates flatpak too, so I’d expect it updates the software
you’ve installed via flatpak. You could test it with the command `flatpak
update`. (Or if it’s installed as the user
nd then migrate the data. Once you have the
correct UUID of the non-LVM volumes updated in your fstab, and the
grub2-efi-x86_64 package installed, you can give it a try to switch over.
--
Jonathan Billings
___
users mailing list -- users@lists.fedorapr
On Jul 8, 2020, at 16:34, Patrick Dupre wrote:
>
> Why cron.weekly does not show up?
Look in /etc/anacrontab.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to us
my disks the last time I installed.
It shouldn’t be difficult to mount the old disk and copy over relevant data.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedorap
On Jul 8, 2020, at 18:42, Ed Greshko wrote:
> So, you think he meant fat or ntfs? I can't recall an "msdos" file system.
No one in this thread is talking about filesystems, but instead disk petition
tables.
Linux and Windows On x86 use disks that are partitioned with a DOS partition
table (
On Jul 8, 2020, at 20:20, Jonathan Billings wrote:
>
> petition tables
Er partition tables.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
is local to the system and not a remote filesystem or
something unusual? I’ll assume it is a normal Linux filesystem.
You might want to make sure the selinux attributes are correct on /home.
Run ‘restorecon -r -v /home’ to fix.
--
Jonathan Billings
__
you shouldn't expect to see anything the Mac
advertises via mdns (i.e. Bonjour, Rendezvous, DNS-SD, zeroconf,
etc. etc.).
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@list
you've mentioned whether a standard Fedora
install on a USB disk would help.
Just booting from the LiveCD with additional kernel parameters would
probably be sufficient to test your system. Can you not interact with
the bootloader of the LiveCD? You could potentially edit the livecd
image and
rtal
> (rw,nosuid,nodev,relatime,user_id=0,group_id=0)
>
> More nonsense from portal and flatpak being dragged in?
I was under the impression that the xdg-document-portal only launched
via dbus when running a graphical session. Are you logging into a
graphical session
On Aug 1, 2020, at 10:14, Patrick Dupre wrote:
> Can we use the GPU with fedora?
> I currently use OpenMP to parallelize my code.
Which GPU? OpenMP is used on many HPC systems with nvidia compute cards, and
some high end workstation GPUs can be used too, I think.
--
Jonathan Bi
the device
(i.e. /dev/sdX ) that was mounted. You need to use the root device
and not a partition.
Also, livecd-iso-to-disk is for putting a livecd image on a flash
drive. As far as I know, Windows 10 isn't a Linux liveCD. I imagine
it will fail
ing
the image to the flash drive. There's a version for Windows and macOS
too.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
h
On Aug 26, 2020, at 20:31, Samuel Sieb wrote:
>
> Did you read the description of the project? It's just a wrapper for the
> Teams web interface, not really an application.
Unfortunately, that description applies to a good many apps these days.
--
Jon
xterm: DISPLAY is not set
What about running xterm on the local display, not after ssh'ing into
localhost? Of course nothing X is going to work after running 'ssh
localhost', you didn't tell it to forward X connections.
--
Jonathan Billings
, keeping the Fn pressed down) should activate and
deactivate the camera.
You can test this by pressing the other hardware buttons like the key for
keyboard backlight, screen brightness and volume, which are activated with Fn
and the marked F keys.
--
Jonathan
On Sep 5, 2020, at 12:09, Patrick Dupre wrote:
>
> Hello,
>
> I have no Wi-fi adapter found.
>
> It is an
> Intel® Dual Band Wireless-AC 3165
>
> This device used to run properly in the past.
> The driver (I guess that the correct)
> iwl7260-firmware-25.30.13.0-111.fc32.noarch
> is installed
0-p-176-iso8859-1
So, I suspect the config for the HP48 emulator (another old memory,
RPN rocks!) has a size 12 font set someplace.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le..
actually mounted. What does “mount
-a” return?
Does /media/nfs exist? Once you get the fstab fixed, test to see if it
automounts.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to
On Sep 13, 2020, at 14:16, Bob Goodwin wrote:
>
> The more I look at this the more confused I get.
Maybe you can explain what you expect to see from the output of “ll /“ on the
NFS server?
--
Jonathan Billings
___
users mailing list --
it
would move the contents of /home/bobg/Public/nfs4exports/home/ into a directory
called /nfs4exports/. So, if there was a
/home/bobg/Public/nfs4exports/home/foo, it would now be /nfs4exports/foo.
You could either ‘mkdir /nfs4exports’ first, or move
/home/bobg/Public/nfs4exports instead.
--
qa.fedoraproject.org/blockerbugs/milestone/33/beta/buglist
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproject.
On Wed, Sep 23, 2020 at 11:01:36AM -0500, David wrote:
> https://communityblog.fedoraproject.org/?p=9285
Returns a 404 for me.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users
n, which supports firewalld, uses ipsets or nftables, and is
considerably faster. It supports reading the journal too so if you don’t want
to use syslogd or send your log traffic off-host, it still works. Since you
have vsftpd, it can also logs for it too.
--
Jonathan Billings
n to debug filesystem behavior, just
poke around /sys/kernel/debug/tracing/events/.
https://www.kernel.org/doc/html/v5.8/trace/events.html
From my experience, as an end user providing feedback, the kernel
developers really like it when you can point out a particular syscall
from the trace out
e leave me a feedback on your experience.
Why are you advertising another distro on a list for Fedora users? Or
are you just spamming mailing lists? I see you also posted to a
debian list too. It doesn't really put a very good light on the
project you are a
Then I googled for the project and discovered it was a package manager
for another distro. Maybe you can describe in a short paragraph why
this is something Fedora users might be interested in? So we don't
have to watch a video of questionable qual
If so, it appears there's an issue with the selinux policy on your
system.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
ht
ing
is probably what is causing the system to appear slow.
Make your terminals wider, top will show more command line info when you press
the ‘c’ key.
You also appear to be running a WINE executable (word pro?) which is also going
to be pretty inefficient.
--
Jonatha
hones/iPads, whatever, (they do not use NFS) can connect to the wifi,
> but I can not mount NFS with my wired connection from the desktop.
When you got the new device, it looks like the local network addressing
changed. Did your NFS client’s IPs change too? Do you need to update
On Oct 16, 2020, at 18:43, Samuel Sieb wrote:
>
> Then how did you get a volume with that name? What is it?
I believe that’s the name of the volume group when you install from a Live
Fedora disk.
--
Jonathan Billings
___
users mailin
ice.html
You can append fsck.mode=force to force a fsck on the next boot for all file
systems.
You could also use tune2fs to set the max mount count to 1, assuming ext4, and
it’ll fsck it next time it tries to mount it. Not sure if there is an
equivalent XFS settin
On Oct 19, 2020, at 16:15, ToddAndMargo via users
wrote:
> Combining excludepkgs, using a space as the delimiter, now
According to the documentation:
exclude
list
Exclude packages of this repository, specified by a name or a glob and
separated by a comma, from all operations. Can be disabl
On Oct 19, 2020, at 17:58, Jonathan Billings wrote:
>
>
>> On Oct 19, 2020, at 16:15, ToddAndMargo via users
>> wrote:
>>
>> Combining excludepkgs, using a space as the delimiter, now
>
> According to the documentation:
>
>
> exclude
anything different than dd, it just
writes the ISO to the disk. It doesn't change the partition scheme,
it doesn't affect the install, it just puts the ISO image of the
Fedora installer on the removable disk.
--
Jonathan Billings
___
nclude ‘conv=sync’ in your dd command if you want to ensure that
everything is written, but it might not as fast, depending on the block size.
I tend to just run ‘eject’ on the USB device, which also prompts the kernel to
write the buffers to disk, but just for that device.
drive
where you can choose to partition your disk and install Fedora.
I've seen several posts trying to explain that Media Writer doesn't
affect your partitioning, but you continue to attribute partitioning
to Media Writer, so I hope this explains things better.
--
Jonathan Billings
sked the packagekit service since I don't want to use
it. I don't see them anymore.
I see there was a bugzilla about this a while ago:
https://bugzilla.redhat.com/show_bug.cgi?id=1382335
I wonder if something returned?
--
Jonathan Billings
___
he X server out
of the conversation, which can improve security and efficiency, and
also makes implimentation simpler.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.
he same issues and not fail.
This is just FUD and ignorant of the Wayland project. Most of the
developers are current X.org developers or have worked on related code
used by X.org (input, drm, etc.). They're well aware of the current
X.org code and limitations in Wayland.
--
Jonathan Billi
should be an opt in and not the default permissions.
--
Jonathan Billings
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproje
yout for Fedora *Server* 32 was
LVM with only a 15G root filesystem, regardless of how much space was
available. The expectation is that the user will expand the root LV or create
other file systems to use the rest of the space free in the VG.
--
Jo
nfiguration, and at first I tried to get the
dnsmasq to be used by systemd-resolved but it kept "forgetting" it and
switching back to what DHCP used, so I just stopped, disabled and
masked systemd-resolved.service, deleted /etc/resolv.conf and
restarted NetworkManager to ge
On Tue, Nov 03, 2020 at 09:08:04AM -0600, Ian Pilcher wrote:
> On 11/2/20 10:37 AM, Jonathan Billings wrote:
> > I have a similar configuration, and at first I tried to get the
> > dnsmasq to be used by systemd-resolved but it kept "forgetting" it and
> > switchi
1 - 100 of 476 matches
Mail list logo
