Re: F40 hibernation issue

[Patrick O'Callaghan]

On Tue, 2024-04-30 at 15:36 -0700, Samuel Sieb wrote:
> > 
> > And the system is not hibernating.
> >    
> > /dev/zram0 exists and is configured:
> 
> Yes, and you can see it being disabled in the logs because you can't 
> hibernate to RAM.
> 
> > # swapon
> > NAME   TYPE  SIZE USED PRIO
> > /SWAP/swapfile file   48G   0B   -2
> > /dev/zram0 partition   8G 512K  100
> 
> Do you have secure boot enabled?

No, however today's log is different:

May 01 01:02:44 Bree systemd[1]: Starting hibernate-preparation.service - 
Enable swap file and disable zram before hibernate...
May 01 01:02:44 Bree kernel: [drm] scheduler comp_1.1.1 is not ready, skipping
May 01 01:02:44 Bree kernel: [drm] scheduler comp_1.0.1 is not ready, skipping
May 01 01:02:44 Bree kernel: [drm] scheduler comp_1.1.1 is not ready, skipping
May 01 01:02:44 Bree kernel: [drm] scheduler comp_1.0.1 is not ready, skipping
May 01 01:02:45 Bree systemd[1]: dev-zram0.swap: Deactivated successfully.
May 01 01:02:45 Bree systemd[1]: hibernate-preparation.service: Deactivated 
successfully.
May 01 01:02:45 Bree systemd[1]: Finished hibernate-preparation.service - 
Enable swap file and disable zram before hibernate.
May 01 01:02:45 Bree systemd[1]: hibernate-preparation.service: Consumed 1.158s 
CPU time.
May 01 01:02:45 Bree systemd[1]: Starting systemd-hibernate.service - System 
Hibernate...
May 01 01:02:45 Bree systemd[1]: Stopping systemd-zram-setup@zram0.service - 
Create swap on /dev/zram0...
May 01 01:02:45 Bree kernel: [drm] scheduler comp_1.1.1 is not ready, skipping
May 01 01:02:45 Bree kernel: zram0: detected capacity change from 16777216 to 0
May 01 01:02:45 Bree kernel: hpet: Lost 1 RTC interrupts
May 01 01:02:45 Bree systemd-sleep[130021]: Performing sleep operation 
'hibernate'...
May 01 01:02:45 Bree kernel: PM: hibernation: hibernation entry
May 01 01:02:45 Bree systemd[1]: systemd-zram-setup@zram0.service: Deactivated 
successfully.
May 01 01:02:45 Bree systemd[1]: Stopped systemd-zram-setup@zram0.service - 
Create swap on /dev/zram0.
May 01 08:00:48 Bree kernel: Filesystems sync: 0.164 seconds
May 01 08:00:50 Bree kernel: Freezing user space processes

and hibernation did work. Possibly a timing issue, so I'll have to look
at my hibernate-preparation script:

$ cat /etc/systemd/system/hibernate-preparation.service
   [Unit]
   Description=Enable swap file and disable zram before hibernate
   Before=systemd-hibernate.service
   
   [Service]
   SyslogIdentifier=%N
   User=root
   Type=oneshot
   
   ExecStart=/bin/bash -c ' \
   swapon -a; \
   swapoff /dev/zram0; \
   exit 0; \
   '
   
   [Install]
   WantedBy=systemd-hibernate.service
   
poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

F40 hibernation issue

[Patrick O'Callaghan]

I've changed nothing since upgrading to F40 a week ago, yet I now find
these errors in the journal:

Apr 30 01:02:39 Bree systemd[1]: Reached target sleep.target - Sleep.
Apr 30 01:02:39 Bree systemd[1]: Starting hibernate-preparation.service - 
Enable swap file and disable zram before hibernate...
Apr 30 01:02:42 Bree systemd[1]: dev-zram0.swap: Deactivated successfully.
Apr 30 01:02:42 Bree systemd[1]: hibernate-preparation.service: Deactivated 
successfully.
Apr 30 01:02:42 Bree systemd[1]: Finished hibernate-preparation.service - 
Enable swap file and disable zram before hibernate.
Apr 30 01:02:42 Bree systemd[1]: hibernate-preparation.service: Consumed 2.698s 
CPU time.
Apr 30 01:02:42 Bree systemd[1]: Starting systemd-hibernate.service - System 
Hibernate...
Apr 30 01:02:42 Bree systemd[1]: Stopping systemd-zram-setup@zram0.service - 
Create swap on /dev/zram0...
Apr 30 01:02:42 Bree systemd-sleep[90707]: Failed to find location to hibernate 
to: Operation not permitted <--
Apr 30 01:02:42 Bree systemd[1]: systemd-hibernate.service: Main process 
exited, code=exited, status=1/FAILURE
Apr 30 01:02:42 Bree systemd[1]: systemd-hibernate.service: Failed with result 
'exit-code'.
Apr 30 01:02:42 Bree systemd[1]: Failed to start systemd-hibernate.service - 
System Hibernate.
Apr 30 01:02:42 Bree systemd[1]: Dependency failed for hibernate.target - 
System Hibernation.
Apr 30 01:02:42 Bree systemd[1]: hibernate.target: Job hibernate.target/start 
failed with result 'dependency'.
Apr 30 01:02:42 Bree systemd-logind[1061]: Operation 'hibernate' finished.
...skipping...

And the system is not hibernating.
 
/dev/zram0 exists and is configured:

# swapon
NAME   TYPE  SIZE USED PRIO
/SWAP/swapfile file   48G   0B   -2
/dev/zram0 partition   8G 512K  100

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: 2 sets of backups with Backups?

[Patrick O'Callaghan]

On Tue, 2024-04-30 at 16:44 +0700, Frederic Muller wrote:
> Hi!
> 
> I currently do weekly backups with Backups and Duplicity> I was
> thinking 
> to add another batch job of daily backup, faster, for specific files
> and 
> folders that are updated daily.
> 
> Unfortunately Backups doesn't seem to give the options for 2 batches,
> or 
> selecting single files. How would you do about that then?
> 
> I have a NAS to which I plan to copy those files.

I use Borgmatic, which runs nightly and has a config file for what I
want to backup and how long to keep it. It's also deduplicating and
optionally encrypts.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: can I clone from a usb flash drive to an NVMe drive?

[Patrick O'Callaghan]

On Tue, 2024-04-30 at 12:17 +0930, Tim via users wrote:
> That's just one example of why I prefer su to sudo, I can easily stay
> in the root environment.  Repeated sudoing can be messy and annoying.

I usually use 'sudo -i' when I need that.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Live USB extra space

[Patrick O'Callaghan]

On Sat, 2024-04-27 at 21:21 -0700, ToddAndMargo via users wrote:
> Hi All,
> 
> I "Finally" have a customer interested in Fedora.  I
> talked him into letting me spin a Live USB flash drive
> for him to play with before we jump ahead.
> 
> Question:  Is there a way to use the extra space on the
> drive to install a few more programs for him to experiment
> with?

Can't you you install them on the running system?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Fedora 40 is available

[Patrick O'Callaghan]

On Fri, 2024-04-26 at 10:00 -0500, Michael Hennebry wrote:
> On Fri, 26 Apr 2024, Patrick O'Callaghan wrote:
> 
> > On Fri, 2024-04-26 at 08:22 +0930, Tim via users wrote:
> > > > as far as xz, for F40, the affected version was only in
> > > > updates-
> > > > testing
> 
> > > Thanks.  Also wondering how far back the breach went.  I remember
> > > reading that the person had being playing the long game about
> > > doing
> > > this.
> > 
> > They spent a long time setting it up, but the actual breach wasn't
> > introduced until very recently, and IIRC was caught before it made
> > it
> > to a stable release.
> 
> How was it caught?
> If it's what I read about recently, 'twas a bit of a fluke.

Not quite a fluke, but certainly something to think about:

https://en.wikipedia.org/wiki/XZ_Utils_backdoor

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Fedora 40 is available

[Patrick O'Callaghan]

On Fri, 2024-04-26 at 08:22 +0930, Tim via users wrote:
> > as far as xz, for F40, the affected version was only in updates-
> > testing
> > for a short time, it never reached stable. The downgrade to the
> > older version was pushed later in the day that the announcement
> > came out
> > about it. 
> 
> Thanks.  Also wondering how far back the breach went.  I remember
> reading that the person had being playing the long game about doing
> this.

They spent a long time setting it up, but the actual breach wasn't
introduced until very recently, and IIRC was caught before it made it
to a stable release.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Fedora 40 is available

[Patrick O'Callaghan]

On Thu, 2024-04-25 at 23:09 +0930, Tim via users wrote:
> On Thu, 2024-04-25 at 11:51 +0100, Patrick O'Callaghan wrote:
> > I've updated to F40. Everything works as expected, (I don't have an
> > Nvidia card at present so YMMV). I do have some comments on the new
> > KDE/Plasma but I'll post on the Fedora KDE list after a bit more
> > testing.
> 
> Did that xv security issue get sorted out before the release?

AFAIK that was fixed almost immediately after it was announced.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Fedora 40 is available

[Patrick O'Callaghan]

On Wed, 2024-04-24 at 22:48 -0400, Jeffrey Walton wrote:
> It looks like Fedora 40 just dropped! Upgrade instructions at <
> https://docs.fedoraproject.org/en-US/quick-docs/upgrading-fedora-offline/
> >.
> 

It was on the Fedora Announce list yesterday. Surprised it didn't
appear here as well as it usually does.

> RPMFushion folks should wait about a month before upgrading. There
> always
> seems to be a lot of video problems after a new release while
> RPMFusion is
> rebuilding their packages.

I've updated to F40. Everything works as expected, (I don't have an
Nvidia card at present so YMMV). I do have some comments on the new
KDE/Plasma but I'll post on the Fedora KDE list after a bit more
testing.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: The best way to still use Fedora + Xorg + Gnome ... even after version 40

[Patrick O'Callaghan]

On Wed, 2024-04-17 at 14:00 +0200, Dario Lesca wrote:
> Ok my notebook I work with Fedora from Core 1 with Gnome on Xorg-x11
> 
> I would like to continue to work with Gnome on Xorg-x11 ... even
> after
> version 40
> 
> What is the best way to do that without change distro?

As I understand it, X11 will still be installable, though I don't know
how long that will last.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: how long does dnf system-upgrade take?

[Patrick O'Callaghan]

On Sat, 2024-04-13 at 17:37 -0400, Fulko Hew wrote:
> > On Sat, 2024-04-13 at 13:29 -0700, Fulko Hew wrote:
> > > I don't update my Fedora systems unless I have a real need,
> > > so yesterday I started updating from F35 to F38

Just as an aside: leaving a two year old Fedora system without updating
is definitely not recommended. F35, F36 and F37 are all EOLed and don't
get even critical security updates. With F40 due for release soon, F38
will also fall into that category in a month or two.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: how long does dnf system-upgrade take?

[Patrick O'Callaghan]

On Sat, 2024-04-13 at 13:29 -0700, Fulko Hew wrote:
> I don't update my Fedora systems unless I have a real need,
> so yesterday I started updating from F35 to F38
> I used the:
> 
> dnf system-update download --releasever=xx
> process.
> 
> - My first update from F35 to F36 worked.
> - Then I did the F36 to F37 update.
>   KDE wouldn't work until I uninstalled qt5-qtwebengine-freeworld
>   to get KDE on X11 to work again.
> - Once that was working I started the F37 to F38
> 
> It did all the downloads, transaction checking, and (presumably)
> installing
> and is now showing:
> 
> Booting 'Fedora Linux (6.8.4-100.fc38.x86_64) 38 (KDE Plasma)'
> 
> It's been about 2 hours... How long do I have to wait?
> It's also too bad, it doesn't tell you what it is or was last doing.
> 
> Help!

Hit Escape during the boot process to get the console log.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: failing https://lists.fedorahosted.org/admin/lists/freeotp-devel.lists.fedorahosted.org/

[Patrick O'Callaghan]

On Mon, 2024-04-08 at 17:28 +0200, Walter H. via users wrote:
> On 08.04.2024 15:32, Michael D. Setzer II via users wrote:
> > Checked
> > https://www.isitdownrightnow.com/lists.fedorahosted.org.html
> 
> answer failed
> 
> the URL in question is
> 
> https://lists.fedorahosted.org/admin/lists/freeotp-devel.lists.fedorahosted.org/
> 

Works for me (though it's slow to appear). I didn't try logging in.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: [OT] plasma6

[Patrick O'Callaghan]

On Thu, 2024-04-04 at 07:28 -0700, Mike Wright wrote:
> On 4/3/24 22:52, Barry wrote:
> > 
> > 
> > > On 3 Apr 2024, at 23:13, Mike Wright
> > >  wrote:
> > > 
> > > If anybody is curious about KDE's new Plasma6 on Wayland you can
> > > install Fedora-KDE-Live from rawhide to get a sneak-peak at
> > > Fedora 41.  Spoiler alert: it's nice.
> > 
> > It is already part of fedora 40.
> > What is it that the rawhide builds providing that is not in f40?
> > 
> Thanks for the heads up.
> 
> KDE isn't part of my bailiwick but a developer waxing enthusiastic 
> convinced me to check it out.  I guess it has undergone a decade long
> upgrade and the change to QT6 required some major rewrites. 
> Possibility 
> of another environment than X11/Gnome was also a factor so I decided
> to 
> try the Kool Aid.
> 
> I've been looking for *any* distro that had picked it up already.  It
> hasn't gained much traction from what I could find.  I had no idea
> that 
> fedora had quietly adopted it and it was sitting there under my nose.
> :/

You haven't been paying attention :-) This has been repeatedly
discussed on the Fedora KDE list, where these things are usually to be
found. The adoption of KDE6 in F40, particularly the demotion of X11 as
default in favour of Wayland, has been a favourite topic.

KDE6 was only finalised recently, so I expect distros to come online in
due course. Fedora being bleeding-edge, it's of course one of the
first.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: shutdown as a simple user

[Patrick O'Callaghan]

On Mon, 2024-04-01 at 11:32 +0200, François Patte wrote:
> Le 2024-04-01 10:40, Samuel Sieb a écrit :
> > On 4/1/24 01:08, François Patte wrote:
> > > Up to yesterday, I was allowed to shutdown and hibernate my
> > > machine as 
> > > a simple user.
> > 
> > Using the gui or the command line?
> 
> I use xfce and in the panel there is tab with the user name, using
> this 
> you can shutdown, hibernate or lock the session. Now shutdown and 
> hibernate are greyed and I can't use them.
> I have to become root in a terminal and use the command init 0.
> 
> > 
> > > Yesterday a powercut while the machine was up suppressed this 
> > > feature I don't understand why.
> > > 
> > > How can recover this feature?
> > 
> > How are you trying to do it and what happens?
> 
> I have no idea about the way to recover... That's why I am asking
> this 
> question.

Crtl-Alt-Del should give you a selection of options including logout
and restart (at least I assume so, though I haven't used XFCE).

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Uh-oh...

[Patrick O'Callaghan]

On Sat, 2024-03-30 at 12:08 -0500, Dave Ihnat wrote:
> Didn't see this go by, but it looks hot enough to risk a repeat
> posting.
> From a friend:
> 
>   It appears there's been a very serious effort to backdoor sshd on
>   Linux via the xz compression/decompression system.
> 
>   https://www.openwall.com/lists/oss-security/2024/03/29/4
> 
>   If you have anything running very recent Linux, it's worth
> investigating
>   whether you're affected.

AFAIK this only applies to Rawhide and the (as yet unreleased) F40,
both of which I assume will be patched ASAP.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: VDQ finding dnf names

[Patrick O'Callaghan]

On Tue, 2024-03-26 at 15:31 -0400, Jon LaBadie wrote:
> On Tue, Mar 26, 2024 at 04:41:33PM -, Beartooth wrote:
> > 
> > I normally have two or three browsers open. Most have been
> > installed for so long I no longer remember what dnf calls them.
> > 
> > My wife, who also runs F39, has been having trouble with
> > the ones
> > she has, and I have yet to get around to trying to diagnose them.
> > Today as
> > a stopgap I tried to install some others for her, using dnf.
> > 
> > It couldn't find brave, falcon, opera, nor vivaldi. VERY
> > Dumb
> > Question : How do I look up what it calls them?
> 
> If you know the path to the browser, perhaps /usr/bin/opera, AND have
> the repo for it active, you could try the dnf command "provides".
> 
> $ dnf provides /usr/bin/opera

A while back I wrote a little script:
   $ cat bin/whats
   #!/bin/bash
   # Identify what package $1 is from, if any
   
   if [ $# -ne 1 ]; then
   echo "Usage: whats arg"
   exit
   fi
   if ! bin=$(which "$1" 2> /dev/null); then
   if ! [ -e "$1" ] ; then # Maybe it's a file
   echo "$1" not found
   exit
   fi
   bin="$1"
   fi
   cand=$(rpm -qf "$bin" 2> /dev/null)
   if echo "$cand" | grep "not owned" > /dev/null 2>&1; then
   file "$bin"
   #   echo No idea
   exit
   fi
   rpm -qi "$cand"

So 'whats opera' would give the package info for the command, assuming
it's installed.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: VDQ finding dnf names

[Patrick O'Callaghan]

On Tue, 2024-03-26 at 20:21 +, Beartooth wrote:
> On Tue, 26 Mar 2024 15:31:30 -0400, Jon LaBadie wrote:
> 
> > On Tue, Mar 26, 2024 at 04:41:33PM -, I Beartooth wrote:
> 
> > >   It couldn't find brave, falcon, opera, nor vivaldi. VERY
> > > Dumb
> > > Question : How do I look up what it calls them?
> > 
> > If you know the path to the browser, perhaps /usr/bin/opera, AND
> > have
> > the repo for it active, you could try the dnf command "provides".
> > 
> > $ dnf provides /usr/bin/opera
> 
>   I tried "whatprovides" on her machine (and got nothing); is
> this a 
> new version of it??

'--whatprovides' is an option to the 'rpm' command, not 'dnf.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Fedora 40 Beta - Gnome Remote Desktop RDP

[Patrick O'Callaghan]

On Tue, 2024-03-26 at 09:56 +0100, Winfried de Heiden wrote:
> Most likely, today Fedora 40 Beta will be released. Well, I couldn't
> wait, so I already downloaded it yesterday. (yes you can) 
> 
> Fedora 40 will include Gnome 46 with a cool new feature: Remote Login
> with RDP. (https://release.gnome.org/46/)  
> 
> However, it does not seem to work in Fedora 40-Beta; I guess because
> there is still an old Gnome 45 packakge included: 
> 
> cat /etc/redhat-release; rpm -qi gnome-remote-desktop 
> 
> Fedora release 40 (Forty) 

Since it's not yet released, the place to post this is the Fedora Test
list. That's where testers hang out, while at least some do not read
this list.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: HP elitedesk 705 G1 mt will not boot from dvd

[Patrick O'Callaghan]

On Mon, 2024-03-25 at 20:57 -0600, Joe Zeff wrote:
> On 03/25/2024 07:57 PM, Tim via users wrote:
> > Brute force and ignorance is a tried and tested method.  Trying to
> > be
> > clever with boot menus, and carefully selecting specific partitions
> > while installing, often goes awry.  Not to mention the times you
> > come
> > across an installer that only wants to do a full takeover of your
> > install drive and nuke everything that was on it.
> 
> In all the years I've been doing this I've never had it fail with a 
> Fedora re-installation.  Of course, I always have a full backup of
> /home 
> before I upgrade or re-install, JIK.

Ditto.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: HP elitedesk 705 G1 mt will not boot from dvd

[Patrick O'Callaghan]

On Mon, 2024-03-25 at 12:36 -0500, Thomas Cameron wrote:
> On 3/25/24 11:38, Patrick O'Callaghan wrote:
> > On Mon, 2024-03-25 at 11:07 -0500, Thomas Cameron wrote:
> > > dmesg > /dev/nvme1n1
> > 
> > What's that about?
> 
> If you overwrite the first few sectors of the drive (i.e. with the 
> output of the dmesg command), it nukes the boot instructions. It
> causes 
> my machine to PXE boot and I kickstart it.
> 
> Sorry, I should have been more clear about that. It could be
> anything. I 
> used to use dd if=/dev/zero of=/dev/nvme1n1 bs=1M count=1 but dmesg
> is 
> quicker to type.

That's what I thought.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: HP elitedesk 705 G1 mt will not boot from dvd

[Patrick O'Callaghan]

On Mon, 2024-03-25 at 11:07 -0500, Thomas Cameron wrote:
> dmesg > /dev/nvme1n1

What's that about?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: change OTP password ...

[Patrick O'Callaghan]

On Sun, 2024-03-24 at 14:05 +0200, Cătălin George Feștilă wrote:
> I asked users if is a problem with logging into the Fedora website
> based on
> new OTP features.
> I fix this issue ... Thank you.

I don't mean to prolong this, but your original post says nothing about
the Fedora website. I notice that you sent it on HyperKitty, which
perhaps made you think it was obvious, but for those of us who only
read the mailing list (I suspect the majority) it wasn't.

Let's leave it at that.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: change OTP password ...

[Patrick O'Callaghan]

On Sun, 2024-03-24 at 01:27 +0100, Jon Ingason via users wrote:
> Den 2024-03-23 kl. 23:28, skrev Patrick O'Callaghan:
> > On Sun, 2024-03-24 at 00:17 +0200, Cătălin George Feștilă wrote:
> > > OTP is the new feature set on login for users ...
> > 
> > I still have no idea what this is about, but never mind.
> > 
> > poc
> > --
> 
> Google otp meaning
> 
> OTP means One Time Password: it's a temporary, secure PIN-code sent
> to 
> you via SMS or e-mail that is valid only for one session. Smart-ID
> uses 
> OTPs during registration and account renewal to confirm your contact 
> information.

I know what an OTP is. My question is what this has to to with logging
into Fedora, to which the answer is 'nothing'. I'm now assuming you're
actually asking about a Fedora-related website, but that wasn't clear
from your original question.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: change OTP password ...

[Patrick O'Callaghan]

On Sun, 2024-03-24 at 00:17 +0200, Cătălin George Feștilă wrote:
> OTP is the new feature set on login for users ...

I still have no idea what this is about, but never mind.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: change OTP password ...

[Patrick O'Callaghan]

On Fri, 2024-03-22 at 19:37 +, Cătălin George Feștilă wrote:
> I just sent an email to reset my OTP, somehow I lost it when I tried
> to change the password.
> Can you tell me if this works now and how fast can be done?
> This is my mail to admin, thank you:
> 
> Dear team 
> 
> I lost my OTP, I tried to change the password but not work.
> Can you reset it and tell me what can I do to recover the account?
> Thank you. Best regards.

OTP for what exactly? Why are you sending this to the Fedora Users
list?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Thu, 2024-03-21 at 21:45 -0400, Charles Dennett wrote:
> 
> 
> On 3/21/24 17:36, Sam Varshavchik wrote:
> 
> > 
> > I keep reading good things said about Brother scanners/all-in-ones.
> > Reading into details, it's always one of two things. It also
> > involves a 
> > driver download, like hplip's modus operandi, and it's not really
> > an out 
> > of the box solution. Alternatively, the alleged printer model is
> > nowhere 
> > to be found in xsane's hardware database. So, it's basically pot
> > luck: 
> > buy it, and hope that it works.
> 
> I've been following this discussion because I currently have an HP-
> 4650 
> 3 in 1 inkjet and I'd like to replace it with a laser 3 in 1.  I've
> been 
> looking at the Brother DCP-L2640DW.  Right now it goes for $200. 
> I've 
> been looking to see if it works with Fedora but have not been having 
> much luck other than reviews on Amazon.  Some say it works out of the
> box on Linux and other say they had to download the driver.  I
> believe 
> it's rather new, being released last summer so I imagine it should
> work 
> as both a printer with cups and scanner with xsane.
> 
> Does anyone have any experience with this printer/scanner/copier?

I have the DCP-L2530DW, which I think is very similar (and probably
uses the same onboard software). Works perfectly.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Thu, 2024-03-21 at 18:18 -0400, Jeffrey Walton wrote:
> On Thu, Mar 21, 2024 at 5:36 PM Sam Varshavchik
>  wrote:
> > [...]
> > I keep reading good things said about Brother scanners/all-in-ones.
> > Reading
> > into details, it's always one of two things. It also involves a
> > driver
> > download, like hplip's modus operandi, and it's not really an out
> > of the box
> > solution. Alternatively, the alleged printer model is nowhere to be
> > found in
> > xsane's hardware database. So, it's basically pot luck: buy it, and
> > hope
> > that it works.
> 
> One comment about Brother... My mother uses Linux Mint and has a
> Brother MFC-J870DW All-in-One over USB. My complaint about the
> MFC-J870DW is, it only supplies a 32-bit driver. So we have to run
> the
> i386 multiarch gear on her 64-bit machine.
> 
> The MFC-J870DW is the only thing stopping me from:
> 
>     sudo apt-get purge $(dpkg --get-selections | grep ":i386" | awk
> '{print $1}')
>     sudo dpkg --remove-architecture i386
> 
> In the past we tried to set up the MFC-J870DW without the Brother
> driver. It never worked that way, so we had to install the driver. I
> suspect the problems had to do with the All-in-One connected directly
> to the machine via USB.

This might be useful:

https://wiki.debian.org/CUPSDriverlessPrinting#ippoverusb

It's for Debian but that doesn't seem to matter.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Thu, 2024-03-21 at 17:36 -0400, Sam Varshavchik wrote:
> I keep reading good things said about Brother scanners/all-in-ones.
> Reading  
> into details, it's always one of two things. It also involves a
> driver  
> download, like hplip's modus operandi, and it's not really an out of
> the box  
> solution.

As I already pointed out, my Brother unit did not require a driver
download. One is available, but I don't use it.

> Alternatively, the alleged printer model is nowhere to be found in  
> xsane's hardware database. So, it's basically pot luck: buy it, and
> hope that it works.

As I haven't tested every model of Brother all-in-one I can't guarantee
that they all work, but I'd be surprised is they didn't. This is the
way all such devices will be expected to work in the near future.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Thu, 2024-03-21 at 08:58 -0400, Sam Varshavchik wrote:
> [root@monster ~]# airscan-discover
> [devices]
>   HP LaserJet MFP M234dw (CB1249) = http://192.168.0.19:8080/eSCL/,
> eSCL
>   HP LaserJet MFP M234dw (CB1249) = https://192.168.0.19:443/eSCL/,
> eSCL
>   HP LaserJet MFP M234dw (CB1249) = http://192.168.0.19:53048/, WSD
>   HP LaserJet MFP M234dw (CB1249) =
> http://[fe80::d2ad:8ff:fecb:1249%254]:53048/, WSD
> 
> It's not showing up in airscan-discover.
> 

Try 'sudo avahi-browse -a -t'

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Thu, 2024-03-21 at 08:58 -0400, Sam Varshavchik wrote:
> Patrick O'Callaghan writes:
> 
> > My Brother all-in-one units (I've had two) have never done that.
> > Also,
> > I can get third-party toner cartridges for very reasonable prices.
> 
> Did you get that unit after crossing your fingers, and hoping that it
> works,  
> or did you verify compatibility somehow. I tried reconciling what's
> on the  
> market now versus xsane's hardware compatibility list without much
> luck.

No, I didn't check. I had an earlier Brother unit for years before it
finally gave up the ghost and I got an updated one (didn't even notice
it does double-sided printing until it worked :-). Any modern scanner
should support DNS-SD via Avahi on Linux. Even more so because
Microsoft has announced that they will no longer ship manufacturer
device drivers for printers and scanners, so the market is being pushed
into standardising. I don't know if some special use-cases might
require a device-specific driver, but I don't personally care.

As regards toner compatibility, Amazon has a number of brands that
work. I tend to get special offers of 2 at a time.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Wed, 2024-03-20 at 14:17 +1030, Tim via users wrote:
> On Tue, 2024-03-19 at 09:00 -0500, Thomas Cameron wrote:
> > But I will never buy another HP again. They screw you on the ink,
> > and 
> > they are apparently using DRM so you can't use third party ink.
> > Screw 
> > that. Never again.
> 
> It's a shame we can't override such shenanigans with custom drivers. 
> That and things like scanner/printer combos wasting all your ink when
> you scan to file without any printing.  It's also annoying that it
> wastes several minutes of your time doing that crap before it'll let
> you scan something.  I don't want to ever buy a combo unit again.
> 

My Brother all-in-one units (I've had two) have never done that. Also,
I can get third-party toner cartridges for very reasonable prices.


poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Which scanner/copier do you use with Fedora?

[Patrick O'Callaghan]

On Mon, 2024-03-18 at 20:49 -0400, Sam Varshavchik wrote:
> Which scanner/copier does everyone use, that works out of the box
> with Xsane? 
> Printing would be nice but I already have an HP that does the job for
> me.

Brother DCP-L2539DW all-in-one wireless scanner/printer. Brother do
have binary drivers for this, but I don't use them.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: dmesg suddenly fails when run as normal user on FC39

[Patrick O'Callaghan]

On Thu, 2024-03-14 at 11:54 -0700, Samuel Sieb wrote:
> On 3/14/24 03:33, Patrick O'Callaghan wrote:
> > I think the error message is unhelpful. 'dmesg' could easily detect
> > that it wasn't running as root and say something more meaningful.
> 
> That requires a change to the dmesg program.  The cause of the error
> is a change in the kernel that dmesg doesn't handle.

Exactly, so given that the kernel has changed, a future version of
'dmesg' should reflect that by giving a useful error message.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: dmesg suddenly fails when run as normal user on FC39

[Patrick O'Callaghan]

On Wed, 2024-03-13 at 16:47 -0700, Samuel Sieb wrote:
> On 3/13/24 16:05, Ron Flory via users wrote:
> >   does not happen on FC38, or any prior RedHat/Fedora version since
> > forever.
> > 
> >     dmesg
> >     dmesg: read kernel buffer failed: Operation not permitted
> > 
> >   Userspace scripts (such as used to read pics from cameras &
> > sdcards) 
> > and many progs often use dmesg to detect or identify things like
> > startup 
> > probe info, USB devs, partition numbers etc.
> > 
> >   I worked around this by setting the suid bit of `which dmesg`,
> > but it 
> > would be rude to force everybody to manually do this as part of 
> > post-install cleanup.
> > 
> >   Hopefully an unintended side-effect and not a new "feature" that 
> > wasn't thought through completely.  A web-search suggests
> > debian/ubuntu 
> > may have been doing this for awhile- but we really don't need to be
> > just 
> > like them...    ;)
> 
> This was intentional and there was a thread about this recently, 
> probably on devel or test.  It's considered to be a big security
> issue.
> 
> If you're in the wheel group, you can use "journalctl -k".

I think the error message is unhelpful. 'dmesg' could easily detect
that it wasn't running as root and say something more meaningful.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Tue, 2024-03-12 at 15:36 -0700, Mike Wright wrote:
> On 3/11/24 14:41, Patrick O'Callaghan wrote:
> > On Sun, 2024-03-10 at 16:39 +, Patrick O'Callaghan wrote:
> > > I'd like to play with LXC but I find the docs not very newbie-
> > > friendly.
> > > I'm trying to follow a guide at:
> > > 
> > > https://brandonrozek.com/blog/lxc-fedora-38/
> > > 
> > > (basically because it mentions Fedora). I followed the steps
> > > closely
> > > and rebooted, but I get the following error:
> > > 
> > >     $ systemd-run --unit=my-unit --user --scope -p "Delegate=yes"
> > > --
> > > lxc-start test
> > >     Running scope as unit: my-unit.scope
> > >     lxc-start: test: lxccontainer.c: wait_on_daemonized_start:
> > > 877
> > > Received container state "ABORTING" instead of "RUNNING"
> > >     lxc-start: test: tools/lxc_start.c: main: 306 The container
> > > failed
> > > to start
> > >     lxc-start: test: tools/lxc_start.c: main: 309 To get more
> > > details,
> > > run the container in foreground mode
> > >     lxc-start: test: tools/lxc_start.c: main: 311 Additional
> > > information can be obtained by setting the --logfile and --
> > > logpriority options
> > >     
> > > Any insights would be welcome. (Just in case, I tried running
> > > with
> > > SElinux turned off, but it made no difference.)
> > 
> > One thing: on running lxc-checkconfig I get:
> > ...
> > Cgroup v1 systemd controller: missing
> > Cgroup v1 freezer controller: missing
> > Cgroup ns_cgroup: required
> > 
> > ...
> > 
> > (everything else is OK).
> 
> The first two are irrelevant.  It uses cgroup2 now.  3rd one I don't 
> understand.
> 
> How is your test container working?
> 
It isn't. I still get the same errors on startup.

> I've been playing along over here and now have a container that
> reports 
> "UNPRIVILEGED true" using lxc-ls -f.  It starts and runs but is 
> unusable.  lxc-start -n C1 -F shows the bootup sequence and it is
> full 
> of [FAILED] sections.  root can't even change to /root: permission 
> denied.  Almost everything is owned by 65534:65534.  If I manually
> set 
> an IP and default route I have networking and it uses my DNS
> container 
> successfully.
> 
> I found this: https://brandonrozek.com/blog/lxc-fedora-38/ "Setting
> up 
> unprivileged containers with LXC on Fedora 38" and how to use systemd
> to 
> start and stop the containers.  It works but doesn't solve the other 
> problems I'm seeing.

Yes, I'd found that page a couple of days ago and tried following it.
Same problems as before.

I'm thinking this isn't worth the hassle. My main interest in lxc was
to run a small containerised VPN, but Fedora seems to have much better
support for docker (via podman) so I'll probably concentrate on that.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Mon, 2024-03-11 at 15:08 -0700, Mike Wright wrote:
> Try this ( works with lxc.apparmor.context ).
> 
> lxc.selinux.context = generated
> 

Same result.

> ...and also ~/.local/lxc/default.conf with these contents:
> 
> lxc.include = /etc/lxc/default.conf

The local default.conf was copied from /etc/lxc and edited.

> lxc.idmap = u 0 10 65536
> lxc.idmap = g 0 10 65536
> 

Those lines are already there. This is the complete file:

$ cat ~/.config/lxc/default.conf
lxc.net.0.type = veth
lxc.net.0.link = lxcbr0
lxc.net.0.flags = up
lxc.net.0.hwaddr = 00:16:3e:xx:xx:xx
lxc.idmap = u 0 10 65536
lxc.idmap = g 0 10 65536
lxc.cgroup.devices.allow=a
lxc.selinux.context = generated

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Sun, 2024-03-10 at 16:39 +, Patrick O'Callaghan wrote:
> I'd like to play with LXC but I find the docs not very newbie-
> friendly.
> I'm trying to follow a guide at:
> 
> https://brandonrozek.com/blog/lxc-fedora-38/
> 
> (basically because it mentions Fedora). I followed the steps closely
> and rebooted, but I get the following error:
> 
>    $ systemd-run --unit=my-unit --user --scope -p "Delegate=yes" --
> lxc-start test
>    Running scope as unit: my-unit.scope
>    lxc-start: test: lxccontainer.c: wait_on_daemonized_start: 877
> Received container state "ABORTING" instead of "RUNNING"
>    lxc-start: test: tools/lxc_start.c: main: 306 The container failed
> to start
>    lxc-start: test: tools/lxc_start.c: main: 309 To get more details,
> run the container in foreground mode
>    lxc-start: test: tools/lxc_start.c: main: 311 Additional
> information can be obtained by setting the --logfile and --
> logpriority options
>    
> Any insights would be welcome. (Just in case, I tried running with
> SElinux turned off, but it made no difference.)

One thing: on running lxc-checkconfig I get:
...
Cgroup v1 systemd controller: missing
Cgroup v1 freezer controller: missing
Cgroup ns_cgroup: required

...

(everything else is OK).

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Mon, 2024-03-11 at 09:16 -0700, Mike Wright wrote:
> On 3/11/24 08:41, Mike Wright wrote:
> > On 3/11/24 04:33, Patrick O'Callaghan wrote:
> > > On Sun, 2024-03-10 at 22:56 -0700, Mike Wright wrote:
> > > > On 3/10/24 15:40, Patrick O'Callaghan wrote:
> > > > > On Sun, 2024-03-10 at 11:13 -0700, Mike Wright wrote:
> > > > > > The last two lines are key.  Add these flags: -F -o
> > > > > > logfile.  The
> > > > > > default loglevel is ERROR.  If you want more detail include
> > > > > > -l
> > > > > > LEVEL.
> > > > > > 
> 
> > > AFAIK 'apparmor' is a Ubuntu-ism. Both my system and the system
> > > in the
> > > container are Fedora 39.
> 
> > That one puzzles me, too.  However, there is no corresponding 
> > lxc.selinux setting.  lxc was a canonical/ubuntu creation so my
> > first 
> > reaction was the same as yours but from some of the comments I've
> > read 
> > on https://discuss.linuxcontainers.org ( another forum with *lots*
> > of 
> > empty space ) that it or something like it may have been integrated
> > into 
> > lxc.
> 
> CORRECTION: the above paragraph is wrong.
> 
> A much easier to read manpage:
> 
> https://manpages.ubuntu.com/manpages/bionic/en/man5/lxc.container.conf.5.html
> 

This appears to the manpage at lxc-container.conf(5).

> § SELINUX CONTEXT
> 
> lxc.selinux.context
> 
> Specify the SELinux context under which the container should be run
> or
> unconfined_t. For example
> 
>  lxc.selinux.context = system_u:system_r:lxc_t:s0:c22
> 

I added that to the config file. It made no difference.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: ghost town

[Patrick O'Callaghan]

On Mon, 2024-03-11 at 10:32 -0400, Ranbir wrote:
> On Sun, 2024-03-10 at 09:50 +, Barry wrote:
> > 
> > The majority of user traffic is
> > on https://discussion.fedoraproject.org/ these days.
> 
> Is there proof of this that we could see?
> 

If you mean have I gathered statistics, then the answer is no. For one
thing, the Discourse forum is for all of Gnome, not just Evolution (in
fact I sometimes get posts on other topics I've no interest in). You
register interest in topics (it doesn't really have an exact equivalent
of threads) and then try to figure out the arcane configuration system
so as to limit traffic to those you're interested in. This is a PITA
and doesn't work well. However I've managed to do it and do see
everything related to Evolution on Discourse, plus of course the new
mailing list, and can confirm that in the last 5 days the Evo list has
received to date 23 messages and the Discourse "list" none.

> The web forum is OK. I find it tough to read. There's too much going
> on. Sometimes I just want to see what people are discussing, but it's
> harder to read the topic titles and I don't know why.

Poor design. That's it. I can't be bothered analysing what exactly is
wrong, but that's my take on it.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Sun, 2024-03-10 at 22:56 -0700, Mike Wright wrote:
> On 3/10/24 15:40, Patrick O'Callaghan wrote:
> > On Sun, 2024-03-10 at 11:13 -0700, Mike Wright wrote:
> > > The last two lines are key.  Add these flags: -F -o logfile.  The
> > > default loglevel is ERROR.  If you want more detail include -l
> > > LEVEL.
> > > 
> > > e.g. lxc-start -n containerName -F -o containerName.log -l WARN
> > > 
> > 
> > This is what I get:
> > 
> > $ lxc-start -n test -F -o test.log -l WARN
> 
> > $ cat test.log
> > lxc-start test 20240310223702.913 ERROR    cgfsng -
> > cgroups/cgfsng.c:__cgfsng_delegate_controllers:2921 - Device or
> > resource busy - Could not enable "+cpu +io +memory +pids"
> > controllers in the unified cgroup 9
> > lxc-start test 20240310223702.934 ERROR    cgfsng -
> > cgroups/cgfsng.c:__cgfsng_delegate_controllers:2921 - Device or
> > resource busy - Could not enable "+cpu +io +memory +pids"
> > controllers in the unified cgroup 9
> 
> 
> I use this:
> 
>  lxc.apparmor.profile=unconfined
> 
> which runs the container as root.  That setting doesn't stop you from
> adding profiles.
> 
> 

AFAIK 'apparmor' is a Ubuntu-ism. Both my system and the system in the
container are Fedora 39.

> If I disable that line I get a cgfsng WARN and the container won't 
> start. ( My containers are used for local services so I'm pretty lax 
> about running them as root. They are also heavily firewalled behind a
> router (also a container which starts the firewall then puts an IP on
> the WAN and sets the default route.  Until the router container comes
> up 
> my host has no network connectivity at all) )
> 
> Also, you're remapping IDs.  What happens if you comment those out?
> 

I thought that was part of the whole point of using separate
namespaces.

> My thought here is to get it running as root first then begin the 
> process of securing it as you see fit.
> 
> 
> I don't have this in my configs but I found this:
> 
>  lxc.cgroup.devices.allow=a
> 

I'll try that.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Sun, 2024-03-10 at 19:37 -0700, Mike Wright wrote:
> On 3/10/24 15:40, Patrick O'Callaghan wrote:
> > On Sun, 2024-03-10 at 11:13 -0700, Mike Wright wrote:
> > > The last two lines are key.  Add these flags: -F -o logfile.  The
> > > default loglevel is ERROR.  If you want more detail include -l
> > > LEVEL.
> > > 
> > > e.g. lxc-start -n containerName -F -o containerName.log -l WARN
> > > 
> > 
> > This is what I get:
> > 
> > $ lxc-start -n test -F -o test.log -l WARN
> ...
> > lxc-start test 20240310223702.945 ERROR    start -
> > start.c:print_top_failing_dir:99 - Permission denied - Could not
> > access /home/poc/.local. Please grant it x access, or add an ACL
> > for the container root
> 
> This one looks interesting
> 

$ ls -ld /home/poc/.local
drwx--. 1 poc poc 56 Aug 28  2021 /home/poc/.local

IOW, it already has x access (and lxc-start is not setuid so it's
running as me).

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Configuring LXC containers

[Patrick O'Callaghan]

On Sun, 2024-03-10 at 11:13 -0700, Mike Wright wrote:
> The last two lines are key.  Add these flags: -F -o logfile.  The 
> default loglevel is ERROR.  If you want more detail include -l LEVEL.
> 
> e.g. lxc-start -n containerName -F -o containerName.log -l WARN
> 

This is what I get:

$ lxc-start -n test -F -o test.log -l WARN
lxc-start: test: cgroups/cgfsng.c: __cgfsng_delegate_controllers: 2921 Device 
or resource busy - Could not enable "+cpu +io +memory +pids" controllers in the 
unified cgroup 9
lxc-start: test: cgroups/cgfsng.c: __cgfsng_delegate_controllers: 2921 Device 
or resource busy - Could not enable "+cpu +io +memory +pids" controllers in the 
unified cgroup 9
lxc-start: test: start.c: print_top_failing_dir: 99 Permission denied - Could 
not access /home/poc/.local. Please grant it x access, or add an ACL for the 
container root
lxc-start: test: sync.c: sync_wait: 34 An error occurred in another process 
(expected sequence number 1)
lxc-start: test: start.c: __lxc_start: 2074 Failed to spawn container "test"
lxc-start: test: tools/lxc_start.c: main: 306 The container failed to start
lxc-start: test: tools/lxc_start.c: main: 311 Additional information can be 
obtained by setting the --logfile and --logpriority options
$ cat test.log
lxc-start test 20240310223702.913 ERRORcgfsng - 
cgroups/cgfsng.c:__cgfsng_delegate_controllers:2921 - Device or resource busy - 
Could not enable "+cpu +io +memory +pids" controllers in the unified cgroup 9
lxc-start test 20240310223702.934 ERRORcgfsng - 
cgroups/cgfsng.c:__cgfsng_delegate_controllers:2921 - Device or resource busy - 
Could not enable "+cpu +io +memory +pids" controllers in the unified cgroup 9
lxc-start test 20240310223702.944 WARN cgfsng - 
cgroups/cgfsng.c:fchowmodat:1251 - No such file or directory - Failed to 
fchownat(15, memory.oom.group, 65536, 0, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW )
lxc-start test 20240310223702.944 WARN cgfsng - 
cgroups/cgfsng.c:fchowmodat:1251 - No such file or directory - Failed to 
fchownat(15, memory.reclaim, 65536, 0, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW )
lxc-start test 20240310223702.945 ERRORstart - 
start.c:print_top_failing_dir:99 - Permission denied - Could not access 
/home/poc/.local. Please grant it x access, or add an ACL for the container root
lxc-start test 20240310223702.945 ERRORsync - sync.c:sync_wait:34 - An 
error occurred in another process (expected sequence number 1)
lxc-start test 20240310223702.945 ERRORstart - start.c:__lxc_start:2074 - 
Failed to spawn container "test"
lxc-start test 20240310223702.945 WARN start - start.c:lxc_abort:1039 - No 
such process - Failed to send SIGKILL via pidfd 16 for process 148416
lxc-start test 20240310223702.958 ERRORlxc_start - 
tools/lxc_start.c:main:306 - The container failed to start

I guess the first two lines are the key to the problem. My default config file 
is:

$ cat ~/.config/lxc/default.conf
lxc.net.0.type = veth
lxc.net.0.link = lxcbr0
lxc.net.0.flags = up
lxc.net.0.hwaddr = 00:16:3e:xx:xx:xx
lxc.idmap = u 0 10 65536
lxc.idmap = g 0 10 65536

The 'test' container is a Fedora 39 instance.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Fora vs. mailing lists

[Patrick O'Callaghan]

On Sun, 2024-03-10 at 11:40 -0500, Thomas Cameron wrote:
> I love mailing lists. I have filters set up they silently go to the 
> correct mail folder, I can read through them at my leisure, and I
> only 
> have to deal with one client - my mail client. My mail client
> defaults 
> to sane viewing rules, threaded, in the order I prefer. It's the same
> experience across every mailing list I'm a member of. I love that.
> It's 
> very accessibility-friendly.

+1

Not long ago the Gnome Project decided to stop supporting a number of
mailing lists, including the list for Evolution, in favour of an
instance of the Discourse web forum. We were told that a forum could
provide everything a list provided, and you could even interact with it
via email, so clearly it was a Good Thing.

It shortly became clear that the user experience of interacting with
Discourse via email was significantly worse than a traditional mailing
list, so a bunch of us set up a new list
(evolution-us...@lists.osuosl.org) where we continue on our merry way.
The Discourse instance still exists (the two mechanisms are entirely
separate with no cross-posting between them), but traffic on the
mailing list is noticeably greater than on the forum, which I think
says something.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Configuring LXC containers

[Patrick O'Callaghan]

I'd like to play with LXC but I find the docs not very newbie-friendly.
I'm trying to follow a guide at:

https://brandonrozek.com/blog/lxc-fedora-38/

(basically because it mentions Fedora). I followed the steps closely
and rebooted, but I get the following error:

   $ systemd-run --unit=my-unit --user --scope -p "Delegate=yes" -- lxc-start 
test
   Running scope as unit: my-unit.scope
   lxc-start: test: lxccontainer.c: wait_on_daemonized_start: 877 Received 
container state "ABORTING" instead of "RUNNING"
   lxc-start: test: tools/lxc_start.c: main: 306 The container failed to start
   lxc-start: test: tools/lxc_start.c: main: 309 To get more details, run the 
container in foreground mode
   lxc-start: test: tools/lxc_start.c: main: 311 Additional information can be 
obtained by setting the --logfile and --logpriority options
   
Any insights would be welcome. (Just in case, I tried running with
SElinux turned off, but it made no difference.)

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: ghost town

[Patrick O'Callaghan]

On Sun, 2024-03-10 at 02:57 +1030, Tim via users wrote:
> 
> *    * *
>     * **
> some tumbleweeds drift through
> 
> 
> So, is everything working fine, or did a new release break everyone's
> systems so badly they can't email any more?   ;-)

What new release?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Copy broken in recent Firefox?

[Patrick O'Callaghan]

On Fri, 2024-02-23 at 08:38 -0500, Sam Varshavchik wrote:
> Is it just me, or is everyone else no longer able to highlight any
> multi- 
> line chunk of code in Firefox, copy, then paste it somewhere else?
> 
> I did some experiments. I can highlight a portion of a line, and that
> can be  
> pasted in another window. Highlight anything more than a line,
> nothing gets  
> pasted. I tested this on multiple sites, so it's not a single site
> playing  
> games with me. Copy/paste does not work either via the mouse middle
> button  
> click, nor right click's copy popup menu, nor the main menu's
> Edit/Copy/
> 
> So, what's broken:
> 
> 1. Something here.
> 2. Firefox.

Works for me. Firefox on KDE (X11), F39.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Where is the source code for a Koji build?

[Patrick O'Callaghan]

On Thu, 2024-02-22 at 04:13 -0500, Jeffrey Walton wrote:
> Hi Everyone,
> 
> I'm new to working with Fedora's Release Engineering. I ran across a
> package that Failed To Build From Source (FTBS) for Fedora 40 at
> . I have C and
> C++ experience, and I believe I can help.
> 
> I'm having trouble making my way around Koji. I can't find the source
> code for wcm-0.20.0-22.fc39 to repro and fix the bug. (The Source
> field does not provide the source files. After cloning, there's only
> one file called 'dead.package').
> 
> Where does one find the source code? Or where do I find the package
> sources that Fedora used?

Might be more appropriate to ask on the Test list.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: gpg decrypt from a script

[Patrick O'Callaghan]

On Wed, 2024-02-14 at 22:49 -0500, Alex wrote:
> > 
> > 
> > > Also, I'm not new to Linux administration, but somewhat new to
> > > sshpass.
> > > Does it only work with ssh passwords or public keys as well?
> > 
> > You don't need sshpass if you have the private key matching the
> > public
> > key already stored on the server. Just use ssh.
> > 
> 
> This would be for use in a script - I'm assuming someone would use it
> as
> part of an automated backup script or when it's not possible to be at
> the
> keyboard to enter that passphrase.
> 

AFAIK a stored private key works just fine with a script.

> Remember "expect"? It sounds reminiscent of that from the 90s.

'expect' is still available in the standard repo, not that I'm
advocating it instead of sshpass.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: wsdd failed Result: resources

[Patrick O'Callaghan]

On Sun, 2024-02-11 at 19:11 -0500, Tom Horsley wrote:
> What on earth does that mean? Is there a contest to see who can
> generate the least useful error message? :-).
> 

It has nothing on the classic IBM message:

Wrong Error

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: OT: what is an NSC?

[Patrick O'Callaghan]

On Sun, 2024-01-28 at 03:04 -0800, ToddAndMargo via users wrote:
> Hi All,
> 
> Sorry for asking an off topic question here, but
> I know you guys will know the answer off the
> top of your heads.
> 
> What is an "NSC" used in this context?
> 
>     1.3.3 NSCs are installed between all wireless
>     networks and the CDE, regardless of whether
>     the wireless network is a CDE
> 
> CDE is Card Holder Environment.

It might be more understandable if you said where this quote comes
from.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: procmail question

[Patrick O'Callaghan]

On Fri, 2024-01-26 at 09:08 -0600, Thomas Cameron wrote:
> I'm reading articles saying procmail is dangerous and unmaintained 
> (https://anarc.at/blog/2022-03-02-procmail-considered-harmful/).
> 
> I get why a setuid root:mail binary is potentially dangerous, but 
> procmail has been in use for decades and I don't think I've ever
> heard 
> of it being used for an exploit except for way back in 2017 and
> 2014 
> (
> https://www.cvedetails.com/vulnerability-list/vendor_id-225/Procmail.h
> tml). 
> 
> 
> Anyone got any recommendations? I've used procmail for decades. I'm 
> pretty familiar with it. I *can* migrate to sieve, but procmail Just 
> Works(TM), so I'm hesitant.
> 
> Is the risk overblown? We're using Postfix and procmail and it seems
> to 
> be really solid. I am not really looking forward to migrating to
> sieve, 
> so I'd rather just stick with what I know, you know?
> 
> What are your thoughts?

I used procmail for years and never had an issue with it. However I
don't like unmaintained software so removed it when support was
dropped. The problem with Sieve (and several other options) is that
they're server-side, so if your server doesn't support them, and you
don't want to run your own local server (plus e.g. fetchmail) you're
dependent on what your mail provider allows.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Problem with chown

[Patrick O'Callaghan]

On Tue, 2024-01-23 at 11:08 -0500, Robert McBroom via users wrote:
> While installing a virtual machine I get a message that the files
> cannot 
> be read by uid104,gid104 these correspond to qemu:qemu. Tried to use 
> chown to explicitly assign the directories to qemu:qemu but chown 
> doesn't recognize qemu as valid or the uid:gid combination. tried 
> setfacl to add qemu permission to read the files, There were no
> messages 
> but the virtual machine setup still could not read the files.
> 
> Any clues to what is going on?

Please copy-paste the exact commands, with their output.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: How do I read result of a QR Code

[Patrick O'Callaghan]

On Mon, 2024-01-22 at 09:49 -0700, Joe Zeff wrote:
> 
> On 1/22/24 00:30, Tim via users wrote:
> > Then, when it went haywire one day I had to tell telephone support
> > the
> > password to sort things out.  Embarrassing, and quite satisfying at
> > the
> > same time.
> 
> When I was doing tech support, the ID10Ts in IT decided to make our 
> passwords expire after 60 days on services inside the firewall. 

One of my banks still does this. I keep telling them this is outdated
security practice from the 1960's but they don't listen. The upshot is
that they're encouraging people to use weak passwords they can
remember, and then recycle them: password1, password2, password3 etc.

They also won't let me use a password manager on their website, so I
have to copy-paste from a different window or tab.

So many of these rules come from CYA.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: gzip eps

[Patrick O'Callaghan]

On Mon, 2024-01-15 at 16:24 +0100, Patrick Dupre via users wrote:
> Hello,
> 
> I am attaching an eps file

No, you didn't.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: VMware woes

[Patrick O'Callaghan]

On Sun, 2024-01-14 at 17:17 +1030, Tim via users wrote:
> Tim:
> > > Which I received base64 encoded, for reasons unfathomable.
> 
> Dave Close:
> > I wonder what did that. It wasn't sent that way:
> >   Content-type: text/plain; charset="utf-8"
> >   Content-transfer-encoding: 8bit
> 
> These days, 8-bit mail ought to be able to traverse the internet
> unscathed (not need encoding), but so many things decide that they
> better re-encode it for reasons best known only to themselves.  The
> SMTP server you posted through, a list mail server it went into, any
> other servers before you received it...
> 
> But I didn't notice anything in that message that wasn't 7-bit, so
> even
> less need to encode it.

I suspect that saying it's UTF-8 would be enough. I doubt that any mail
server would bother scanning the content to see if it really needed
encoding.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: VMware woes

[Patrick O'Callaghan]

On Sun, 2024-01-14 at 08:29 +1030, Tim via users wrote:
> On Fri, 2024-01-12 at 21:33 -0800, Dave Close wrote:
> > Speaking of email clients modifying received messages ...
> > 
> > {This message was not written in HTML. If you are reading it as
> > such,
> >  the presentation is dishonest and not what the author intended.}
> 
> Which I received base64 encoded, for reasons unfathomable.

AFAIK there's nothing wrong with that. It's plaintext in UTF-8, so
Base64 would be a reasonable encoding.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: errors logged after USB disk ejected

[Patrick O'Callaghan]

On Sat, 2024-01-06 at 00:39 +1100, fed...@eyal.emu.id.au wrote:
> This happens whenever I connect and mount a USB disk
>   $ sudo mount /dev/sdi1 /sata
> Then when I am finished with it, I do
>   $ sudo umount /sata
> and then
>   $ sudo eject /dev/sdi
> At this point the system log shows the below burst of errors (the
> disk is still plugged in).
> 
> Checking the system log I see this for the first time with kernel
> 6.5.12-200.fc38.x86_64
> 
> What is the correct order of commands when removing a USB disk?
> 

AFAIK eject unmounts the drive, so you're attempting to unmount it
twice. For removable USB drives, unmount on its own is enough.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: package/mod install issue

[Patrick O'Callaghan]

On Thu, 2023-12-28 at 06:18 -0500, bruce wrote:
> Working with a test DO process, to install a package/module

I've no idea what "a test DO process" means.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: F38 - ssh now having a password popup

[Patrick O'Callaghan]

On Mon, 2023-12-25 at 22:51 -0500, Robert Moskowitz wrote:
> > The rsync manpage has this suggestion:
> > 
> > set environment variable RSYNC_PASSWORD to the password
> 
> set RSYNC_PASSWORD = paswd
> 
> did not help

You can also use the --password-file option.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: ipp-usb constantly running?

[Patrick O'Callaghan]

On Mon, 2023-12-25 at 14:35 -0500, Tom Horsley wrote:
> I might print something once a week (if that), I've got a USB
> connected
> Brother laser printer. I now see ipp-usb running all the time at
> about
> 3% cpu and using about 2.6 meg of memory. Why is it using cpu all
> the time when I'm almost never printing? If I remove it, will I
> be able to print the old way from before this "improvement"?

I use a Brother laser printer via IPP, but with WiFi and Avahi. As far
as I can tell, nothing is running unless I actually print or scan.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: 6.6.x kernels do not boot successfully

[Patrick O'Callaghan]

On Fri, 2023-12-15 at 13:20 -0500, Terry Polzin wrote:
> 6.6.6 is not booting for me with XFCE but 6.6.4 is.  Why wasn't 666
> skipped
> in sequence? :-)
> 

It's the Kernel of The Beast :-)

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: 6.6.x kernels do not boot successfully

[Patrick O'Callaghan]

On Thu, 2023-12-14 at 23:23 +, Barry wrote:
> 
> 
> > On 14 Dec 2023, at 21:39, Toni Andjelkovic via users
> >  wrote:
> > 
> > Unfortunately, the 6.6.6 release didn't fix anything for me, the
> > boot deadlock
> > persists.
> 
> Are these problems reported in fedora bug? If not then maybe no one
> is trying to fix.

There have been multiple reports across distros. I think upstream has
released 6.6.7, so I guess that will appear in the repo soon.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: I think Firefox crashed my system

[Patrick O'Callaghan]

On Sun, 2023-12-10 at 16:08 -0700, Joe Zeff wrote:
> On 12/10/2023 03:09 PM, Patrick O'Callaghan wrote:
> > In my case at least there wasn't even a mouse cursor on screen, and
> > the
> > fact that nothing was written to the journal for about 10 hours
> > before
> > I rebooted would seem to indicate everything was dead. I should
> > have
> > tried an ssh from my laptop but was too impatient to get the system
> > up.
> 
> Have you tried activating the Magic SysRq key and using it when this 
> happens?

It's never happened before. If it happens again I'll consider it.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: I think Firefox crashed my system

[Patrick O'Callaghan]

On Mon, 2023-12-11 at 07:03 +1030, Tim via users wrote:
> On Sun, 2023-12-10 at 16:50 +0000, Patrick O'Callaghan wrote:
> > It should not be possible for a userland process to cause this,
> > which i
> > why I thought it could be OOM.
> 
> Hmm, I've had a mpv playing a video file hard lock-up a system
> several
> times (dunno if it was a corrupt video file, or another cause). 
> Suddenly the screen would go oddly darker, looking like there was a
> flyscreen in front of the image, then start to smear video as it
> slowly
> drew on the screen.
> 
> I expected to be able to CTRL ALT BACKSPACE to kill X and recover,
> but
> rarely did it recover from this.  The system seems far from crash-
> proof.

In my case at least there wasn't even a mouse cursor on screen, and the
fact that nothing was written to the journal for about 10 hours before
I rebooted would seem to indicate everything was dead. I should have
tried an ssh from my laptop but was too impatient to get the system up.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: I think Firefox crashed my system

[Patrick O'Callaghan]

On Sun, 2023-12-10 at 07:06 -0600, Roger Heflin wrote:
> This is not an out of memory crash.
> 
> This is a code bug in firefox and/or in the shared library being
> called by firefox.
> 
> Usually this is related to something unexpected/unhandled in the data
> stream being processed incorrectly.
> 
> I have had a few of these firefox crashes since around Nov 20.
> 
> Hopefully someone fixes it soon.
> 
I assumed it was an OOM since that's mostly what pops up on Google
searches, but it may not be.

What I neglected to point out was that when I turned on my monitor this
morning the system was completely unresponsive, i.e. no mouse and no
keyboard. The screen showed my (frozen) desktop session. The journal
ended with the crash I reported, i.e. nothing happened after that
including my usual backup and hibernation routines. I had to hard reset
the system to get back in.

It should not be possible for a userland process to cause this, which i
why I thought it could be OOM.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

I think Firefox crashed my system

[Patrick O'Callaghan]

This morning I found this in my journal:

Dec 09 23:53:04 Bree kernel: traps: Isolated Web Co[188480] general protection 
fault ip:7f6a44c75871 sp:7ffc2a7be1f0 error:0 in libxul.so[7f6a42462000+4664000]
Dec 09 23:53:04 Bree systemd[1]: Started systemd-coredump@12-239925-0.service - 
Process Core Dump (PID 239925/UID 0).
Dec 09 23:53:05 Bree abrt-dump-journal-oops[182510]: abrt-dump-journal-oops: 
Found oopses: 1
Dec 09 23:53:05 Bree abrt-dump-journal-oops[182510]: abrt-dump-journal-oops: 
Creating problem directories
Dec 09 23:53:06 Bree abrt-dump-journal-oops[182510]: Reported 1 kernel oopses 
to Abrt
Dec 09 23:53:07 Bree abrt-server[239927]: Can't find a meaningful backtrace for 
hashing in '.'
Dec 09 23:53:07 Bree abrt-server[239927]: Deleting non-reportable oops '.' 
because DropNotReportableOopses is set to 'yes'
Dec 09 23:53:07 Bree abrt-server[239927]: 'post-create' on 
'/var/spool/abrt/oops-2023-12-09-23:53:05-182510-0' exited with 1
Dec 09 23:53:07 Bree abrt-server[239927]: Deleting problem directory 
'/var/spool/abrt/oops-2023-12-09-23:53:05-182510-0'
Dec 09 23:53:07 Bree abrt-server[239927]: Lock file '.lock' was locked by 
process 239940, but it crashed?
Dec 09 23:53:07 Bree systemd-coredump[239926]: [] Process 188480 (Isolated Web 
Co) of user 1000 dumped core.
   
   Module libxcb-dri3.so.0 from rpm 
libxcb-1.13.1-12.fc39.x86_64
   Module libnuma.so.1 from rpm 
numactl-2.0.16-3.fc39.x86_64
   Module libogg.so.0 from rpm 
libogg-1.3.5-6.fc39.x86_64

etc. etc. (I'll omit the rest of the trace)

libxul.so is part of Firefox (and Thunderbird, which I don't use).
Apparently "Isolated Web Co" is a Firefox process which has been
reported for years as leaking memory, though I haven't noticed it doing
this myself (I have a lot of swap). Some people think it's related to
viewing YouTube videos, which I do do.

My question is: why doesn't the OOM killer stamp on this before it gets
out of hand? Isn't that what it's for?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: btrfs Backups

[Patrick O'Callaghan]

On Sat, 2023-12-09 at 11:26 -0500, Tim Evans wrote:
> I mentioned 'rsnapshot' in a previous reply.  While this has similar 
> snaoshotting capabilities, it's a very old package that doesn't have
> a 
> GUI and--most importantly--doesn't incorporate recovery of backed-up 
> files; 'backintime' has both.

IIRC file recovery from an rsnapshot backup consists of mounting the
backup filesystem and copying. Unlike some backup systems (including
Borg) the backup is simply a copy of the file tree with hard links to
reduce the number of physical copies. It's similar to Apple's Time
Machine (though Apple changed normal UNIX/Linux filesystem semantics to
allow hard links to directories, which rsnapshot doesn't do).

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: btrfs Backups

[Patrick O'Callaghan]

On Fri, 2023-12-08 at 12:56 -0500, Tim Evans wrote:
> Since tine immemorial (I first touched a UNIX system circa 1984), I
> have 
> used the venerable 'dump' utility to do automated full and
> incremental 
> backups to an NFS-mounted network storage appliance (NAS).
> 
> Now, with a brand new laptop, with fresh install using btrfs 
> filesystems, I find 'dump' does not work on them. (Fortunately, I was
> able to use the companion utility 'restore' to recover my /home
> backups 
> and assorted config files from the root filesystem of my old laptop.)
> 
> So, I'm looking re-script my backups using btrfs tools.  So far, it 
> appears I cannot do btrfs snapshots to the NAS and I'm not wanting to
> attach another disk to the laptop just for snapshots. Are there 
> analogous (to 'dump' and 'restore') btrfs utilities that can do
> backups 
> this way, and can be scripted?

Whenever someone asks about backup tools, there are usually lots of
answers, so here's mine:

I have BTRFS on my root and /home. My backup target is a USB-attached
dual-slot caddy with a couple of drives in it. They are also BTRFS, and
are formatted RAID-1 in case of hardware issues. My backup tool is
Borg, because a) it's very feature-complete, b) does compression, and
3) does deduplication. I use it via an additional tool called
Borgmatic, which takes charge of scheduling:

$ rpm -qi borgmatic
Name: borgmatic
Version : 1.8.5
Release : 1.fc39
Architecture: noarch
Install Date: Sun 03 Dec 2023 09:43:44 GMT
Group   : Unspecified
Size: 1060831
License : GPL-3.0-or-later
Signature   : RSA/SHA256, Fri 24 Nov 2023 14:30:30 GMT, Key ID 75cf5ac418b8e74c
Source RPM  : borgmatic-1.8.5-1.fc39.src.rpm
Build Date  : Mon 20 Nov 2023 15:54:37 GMT
Build Host  : buildvm-x86-32.iad2.fedoraproject.org
Packager: Fedora Project
Vendor  : Fedora Project
URL : https://torsion.org/borgmatic
Bug URL : https://bugz.fedoraproject.org/borgmatic
Summary : Simple Python wrapper script for borgbackup
Description :
borgmatic (formerly atticmatic) is a simple Python wrapper script for
the Borg backup software that initiates a backup, prunes any old backups
according to a retention policy, and validates backups for consistency.

Also:

https://torsion.org/borgmatic/

Once I had it set up to my liking, I just let it run every night. Every
morning I check (via logs) that it worked without problems. This has
saved me more than once (either from hardware problems or from some
poorly implemented change to my configuration).

I've been generally happy with Borg. My main complaint is that the
config file for Borgmatic is written in YAML, and you are assumed to
know how this works (it's not obvious if you've never used YAML
before).

I used to use rsnapshot, based on rsync. It's somewhat easier to
understand but doesn't do deduplication or compression.

Something people often forget: can you actually restore files from your
backup? Are you sure you know how to do this? Have you written it down
somewhere that won't be lost in the event of disaster?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Thu, 2023-12-07 at 01:21 +1030, Tim via users wrote:
> Tim:
> > > I have one of those two-bay USB - SATA hard drive gadgets, it has
> > > clone drive function that works all by itself.  I've used it a
> > > few
> > > times for relatively pain-free drive cloning.
> 
> 
> Patrick O'Callaghan:
> > I have one of those, which I use in a RAID1 configuration for my
> > Borg
> > backup. However the SSD I wanted to update is a 2TB unit and I
> > don't
> > have a spare of the same size.
> 
> Mine made some claim that you could access both slots from the PC,
> but
> I've never managed it.  I wonder if it required a cloned drive in the
> second slot and *had* to be used as RAID?

I can certainly access both individual drives and can run smartctl on
each of them separately. If it matters, my RAID config is based on
BTRFS and has been working fine.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: f38: 2 Thunderbird misbehaviors.

[Patrick O'Callaghan]

On Wed, 2023-12-06 at 08:57 -0500, John Mellor wrote:
> On 2023-12-05 22:59, home user wrote:
> > . . .
> > I almost always use Gnome with X, not wayland.  Even the test I 
> > reported in my previous post (in response to Richard) was KDE with
> > X, 
> > not wayland.  Should thunderbird-wayland be installed and used by
> > us X 
> > users? 
> X is no longer maintained properly, and by F38 is no longer fully 
> functional.  More functionality removal has happened in F39, and I 
> believe that it is on the feature list of F40 for full removal.  The
> KDE 
> and other desktops are also going away from X for the same reasons.  
> Fedora (and almost every distro now) has gone with Wayland - the 
> successor to X mostly written by the original X developers. So, while
> many parts of the X APIs are working and therefore most of your apps
> are 
> functioning properly, you need to get yourself off of depending upon
> it 
> in the next 6 months or so.  Things like remote X desktops have been 
> designed out of Wayland, so you are maybe going to lose functionality
> and have to use lesser-capable mechanisms like Microsoft-style
> desktop 
> sharing for example.

I'll do that once KDE-Wayland is fully functional. Currently it simply
is not. For one thing, there is no support for session save/restore,
which IMHO is a basic feature.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Wed, 2023-12-06 at 17:05 +1030, Tim via users wrote:
> On Tue, 2023-12-05 at 16:44 +0000, Patrick O'Callaghan wrote:
> > I made some effort to anticipate disaster by copying my /, /boot
> > and
> > /boot/efi partitions to a spare drive (and modifying its /etc/fstab
> > appropriately), but it didn't work. No doubt I would have needed to
> > update the EFI partition, but felt that was a bridge too far and I
> > could in a pinch recover with a Live ISO (all local modifications
> > are
> > backed up anyway).
> 
> I have one of those two-bay USB - SATA hard drive gadgets, it has a
> clone drive function that works all by itself.  I've used it a few
> times for relatively pain-free drive cloning.

I have one of those, which I use in a RAID1 configuration for my Borg
backup. However the SSD I wanted to update is a 2TB unit and I don't
have a spare of the same size.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Wed, 2023-12-06 at 01:29 +1030, Tim via users wrote:
> On Tue, 2023-12-05 at 13:29 +0000, Patrick O'Callaghan wrote:
> > Crossed my fingers and did it. All good.
> 
> Close your fingers and cross your eyes, flash the firmware for a big
> surprise...
> 
> Reminds me that I ought to get a spare drive, I just used my spare
> one
> in something.  When it comes to system updates, and drive firmware is
> in the same boat, I prefer doing a drive swap.  Remove old, fit new,
> apply all the new software, add the old drive and either drag data to
> the new drive, or use it as the data drive.  I completely distrust
> install new releases over old.

I made some effort to anticipate disaster by copying my /, /boot and
/boot/efi partitions to a spare drive (and modifying its /etc/fstab
appropriately), but it didn't work. No doubt I would have needed to
update the EFI partition, but felt that was a bridge too far and I
could in a pinch recover with a Live ISO (all local modifications are
backed up anyway).

Of course if I hadn't gone through all this rigmarole then Murphy's Law
dictates that the firmware update would have wiped the SSD, so I
decided to consider it as a way to propitiate the gods.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Sun, 2023-12-03 at 16:13 +, Patrick O'Callaghan wrote:
> On Sun, 2023-12-03 at 14:38 +0100, Ralf Corsépius wrote:
> > Am 03.12.23 um 13:16 schrieb Patrick O'Callaghan:
> > > I have a 2TB Samsung 860 EVO SSD as my main drive. It's working
> > > well,
> > > but Samsung released updated firmware since I got it, and their
> > > update
> > > tool only works under Windows.
> > 
> > Samsung offers OS-independent firmware ISOs for their consumer
> > SSDs.
> > 
> > Download the ISO, put it on USB-stick, boot from the stick and
> > install 
> > the firmware.
> > 
> > 
> 
> I wasn't aware of that, thanks.

Crossed my fingers and did it. All good.

Thanks again.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Mon, 2023-12-04 at 12:17 +0100, Heinz Diehl wrote:
> > "Will or may delete all the data"? WTF does that mean?
> 
> Exactly that.

"Will or may" is not an intelligible statement. If it definitely will,
than say so. If it's not certain that it will, then say that it may.
Saying both is simply nonsense.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Sun, 2023-12-03 at 14:38 +0100, Ralf Corsépius wrote:
> Samsung offers OS-independent firmware ISOs for their consumer SSDs.

Turns out it's actually a minimalist Linux system, which is hilarious.

> Download the ISO, put it on USB-stick, boot from the stick and
> install the firmware.

I tried it and got the scary message that the process "WILL OR MAY
DELETE ALL THE DATA ON THE DISK".

"Will or may delete all the data"? WTF does that mean?

I do have backups of my data, but what are the chances I'll have to
reinstall the system if I do this?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Sun, 2023-12-03 at 07:22 -0500, Jeffrey Walton wrote:
> On Sun, Dec 3, 2023 at 7:16 AM Patrick O'Callaghan
>  wrote:
> > 
> > I have a 2TB Samsung 860 EVO SSD as my main drive. It's working
> > well,
> > but Samsung released updated firmware since I got it, and their
> > update
> > tool only works under Windows.
> > 
> > Any suggestions on how I could update the SSD? (I know, "if it's
> > not
> > broken, don't fix it" :-)
> 
> I understand some of the EVO's need the firmware update. Otherwise
> they die an early death.
> 
> Give this a try. It is part of my "full update" scripts". It updated
> a
> EVO 970 SSD I installed in a laptop in early October.
> 
> if [[ -n $(command -v fwupdmgr 2>/dev/null) ]]; then
>    if [[ -n $(fwupdmgr get-devices 2>&1 | grep 'UEFI ESRT device')
> ]]; then
>    echo "Updating firmware"
>    fwupdmgr refresh 1>/dev/null && fwupdmgr update 1>/dev/null
>    fi
> fi

It told me updates weren't available for the SSD. Now trying the
Samsung ISO suggested by Ralf.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Samsung SSD firmware

[Patrick O'Callaghan]

On Sun, 2023-12-03 at 14:38 +0100, Ralf Corsépius wrote:
> Am 03.12.23 um 13:16 schrieb Patrick O'Callaghan:
> > I have a 2TB Samsung 860 EVO SSD as my main drive. It's working
> > well,
> > but Samsung released updated firmware since I got it, and their
> > update
> > tool only works under Windows.
> 
> Samsung offers OS-independent firmware ISOs for their consumer SSDs.
> 
> Download the ISO, put it on USB-stick, boot from the stick and
> install 
> the firmware.
> 
> 

I wasn't aware of that, thanks.


poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Samsung SSD firmware

[Patrick O'Callaghan]

I have a 2TB Samsung 860 EVO SSD as my main drive. It's working well,
but Samsung released updated firmware since I got it, and their update
tool only works under Windows.

Any suggestions on how I could update the SSD? (I know, "if it's not
broken, don't fix it" :-)

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: how to remove unwanted old kernels.

[Patrick O'Callaghan]

On Fri, 2023-12-01 at 10:28 -0700, home user wrote:
> On 12/1/23 10:06 AM, Tim via users wrote:
> > On Fri, 2023-12-01 at 09:11 -0700, home user wrote:
> > > What is the current, simple, best practice, f38 way of removing
> > > the
> > > oldest memtest, both from the hard drive (if it's there) and from
> > > the grub menu?
> > 
> > It *ought* to be as simple as
> > 
> > dnf remove kernel.
> > 
> > Where you use a specific kernel package version name.  And it does
> > everything.
> > 
> > Which is pretty much happens when you install a new kernel update
> > and
> > it removed the oldest one, automatically.
> > 
> > You can play with the DNF configuration files to specify how many
> > kernels (install-only packages) to keep on your system, and let it
> > automatically handle this for you.  But I always suggest keeping
> > more
> > than 3, I go for 5.  It might be some time before you notice that a
> > kernel update has caused a problem, and it's handy to be able to go
> > back more than just one, to test things.
> > 
> 
> Because of past problems with hard drive space when doing upgrades,
> I've got it set to keep 3.  (installonly_limit=3).
> How do I get the full "name" of the kernel to be deleted without re-
> booting (so I can see the grub menu)?

'rpm -qa kernel' will list all the installed kernels. The one to be
deleted on the next dnf update will be the oldest.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Upgrade KDE Wayland Spin 38->39 KDE Background Screen Issue

[Patrick O'Callaghan]

On Fri, 2023-12-01 at 05:43 -0800, richard emberson wrote:
> Where do I find "the Fedora KDE list"?

k...@lists.fedoraproject.org

but I see you found it.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Upgrade KDE Wayland Spin 38->39 KDE Background Screen Issue

[Patrick O'Callaghan]

On Thu, 2023-11-30 at 15:16 -0800, richard emberson wrote:
> Upgrade KDE Wayland Spin 38->39 KDE Background Screen Issue
> 
> KDE Screen Issue (or Plasma issue, or...)
> 
> 1) Login screen comes up and can login.
> 2)  KDE screen mix of colors and black... moving mouse and screen
> continues
>    to redraw random colors (but, mostly black).
> 3) Bottom panel (in hide mode) comes up and looks normal. Top ToolBar
>    is drawing like the screen (not normal) but I can tell its being
> drawn
>    ontop of the screen background (different colors/black).
> 5) Firefox comes up rendering like normal.
> 6) Thunderbird asks for password and it comes up drawing like normal.
> 7) Using Mouse Right Button, menu come up (look normal) and can
> 8 "Open Terminal". Kconsole comes up like normal.
> 9) From Kconsole can launch alacritty multiple times all normal.
> 10) Background and ToolBar still not rendering correctly.
> 11) Can move mouse over ToolBar and popups appear. Can also click and
>    launch/hide terminals, firefox, thunderbird when mouse if over
> their
>    position in ToolBar.
> 12) I can render a jpeg photo in eog and it comes up just fine.
> 
> With Fedora 38 KDE Spin I did not have this issue with the background
> or ToolBar.
> So, has anyone seen this behavior?
> Where do I start?
> 

I'd suggest asking on the Fedora KDE list.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Black screen after login to F39

[Patrick O'Callaghan]

On Thu, 2023-11-30 at 09:36 -0800, Michael Eager wrote:
> I upgraded from Fedora 38 to Fedora 39 on an AMD Ryzen 5 5600
> with Nvidia Geforce RTX 3050 graphics card.  I'm running KDE
> Plasma-X11, not Wayland.
> 
> I tried to install the GPU drivers several ways using repos/RPMs,
> but was not successful.  I am using the Nvidia proprietary
> drivers.  They support dual monitors with no problems; I've had
> problems with the nouveau driver and dual monitors in the past.
> 
> I start the system in multi-user (CLI) mode and then run "startx"
> to start X11 and Plasma.  (I started doing this in F38 because of
> sddm hanging, which I was not able to resolve.)
> 
> When I log in with my user account and run startx, I get a black
> screen (extending across both monitors) with an X cursor.  The
> cursor moves in response to the mouse, but neither left or right
> mouse button do anything. I can get to an alternate console
> using alt-F3 and kill the login.
> 
> I have discovered a bit of a rain dance to get a working system.  I
> can log in as root, run startx, and get a working KDE Plasma-X11
> screen.  Then I log out to the sddm screen and log in again with
> my user account.  After this, everything (or almost everything)
> works correctly.
> 
> I've checked the system logs and the Xorg log for any error
> or warning messages, but nothing looks incorrect or at all
> informative.
> 
> Does anyone have any suggestions or insights into this problem?

Not really, but you might want to ask on the Fedora KDE list.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: kernel-6.6.2-201.fc39.x86_64 does not boot, process hangs

[Patrick O'Callaghan]

On Wed, 2023-11-29 at 14:28 +, olivares33561 via users wrote:
> Dear fellow fedora users,
> 
> upon updating and rebooting the new kernel refuses to boot or just
> hangs.  How do I troubleshoot this issue?
> I have to boot 6.5.12-300 so I can use the computer.

Not enough information to go on. What actually happens? Do you see
anything onscreen? What happens if you remove the rhgb parameter from
the boot line?

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Intel NUC 12 Enthusiast Kit with Arc 770 and fedora 39

[Patrick O'Callaghan]

On Tue, 2023-11-21 at 10:08 +0100, Gianluca Cecchi wrote:
> BTW: I have still to understand if the system will use both graphics
> cards
> (which one is primary in this case?) or if the onboard one will be
> disabled
> if using a discrete card on the Nuc.

I've no idea what will happen with a NUC, but in my experience Fedora
will use the discrete card in preference to the IGP, unless it's
blacklisted in the boot line.

There's a package called switcheroo-control which allows you to select
which GPU to use for specific apps, though personally I haven't had
much luck with it.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Sat, 2023-11-18 at 12:59 +, Patrick O'Callaghan wrote:
> On Tue, 2023-11-14 at 13:52 +0000, Patrick O'Callaghan wrote:
> > My system (both F38 and now F39) reboots itself every morning at
> > 8am,
> > yet nothing in the cron configuration is telling it to do that and
> > I
> > don't see anything obvious in the journal to cause it.
> > 
> > How can I figure out what is triggering this? I know the
> > description
> > is
> > vague, but there it is.
> 
> Just to close this off:
> 
> - The problem was being caused by a flaky smart plug
> - Restoring factory settings to the plug fixed it
> - I decided to follow some recommendations in the thread and use the
> built-in motherboard RTC instead of the smart plug as a way to re-
> activate the system at a set time. This is now working (touch wood).
> 
> My hibernation script now includes this snippet, if anyone's
> interested:

Corrected version. It turns out that the wakeup alarm has to be reset
before setting:

set_wakeup() {  # See 
https://www.linux.com/training-tutorials/wake-linux-rtc-alarm-clock/
if [[ $(date +"%H") -lt 8 ]]; then
datestring=$(date '+%s' -d '08:00 today')
else
datestring=$(date '+%s' -d '08:00 tomorrow')
fi
echo 0 > /sys/class/rtc/rtc0/wakealarm
echo "$datestring" > /sys/class/rtc/rtc0/wakealarm
}
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Tue, 2023-11-14 at 13:52 +, Patrick O'Callaghan wrote:
> My system (both F38 and now F39) reboots itself every morning at 8am,
> yet nothing in the cron configuration is telling it to do that and I
> don't see anything obvious in the journal to cause it.
> 
> How can I figure out what is triggering this? I know the description
> is
> vague, but there it is.

Just to close this off:

- The problem was being caused by a flaky smart plug
- Restoring factory settings to the plug fixed it
- I decided to follow some recommendations in the thread and use the
built-in motherboard RTC instead of the smart plug as a way to re-
activate the system at a set time. This is now working (touch wood).

My hibernation script now includes this snippet, if anyone's
interested:

set_wakeup() {  # See 
https://www.linux.com/training-tutorials/wake-linux-rtc-alarm-clock/
if [[ $(date +"%H") -lt 8 ]]; then
datestring=$(date '+%s' -d '08:00 today')
else
datestring=$(date '+%s' -d '08:00 tomorrow')
fi
echo "$datestring" > /sys/class/rtc/rtc0/wakealarm
}

Thanks again to all who chipped in.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: DNF Upgrade from F38 to F39 Issues

[Patrick O'Callaghan]

On Fri, 2023-11-17 at 22:28 -0500, Jeffrey Walton wrote:
> A couple of thoughts in no particular order. I don't know how you
> should move forward with your issues.
> 
> * The KDE spin uses Wayland, not X11.
> 

I use KDE with X11. You just have to select it on the login screen.
AFAIK KDE 6 will default to Wayland in an upcoming release, without
support for X11, but that isn't currently the case.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Fri, 2023-11-17 at 14:50 +1030, Tim via users wrote:
> Roger Heflin wrote:
> > > No, we both mean use the bios timed wake up to wake up your
> > > machine
> > > instead of the plug for anything.
> 
> Patrick O'Callaghan:
> > Ah, I misunderstood. I hadn't realised that was even possible.
> 
> Way back in the day of my brief torture of Win98SE, I used to explore
> the BIOS to see what options there were, and which stupid ones were
> preselected.  Wake on timer has been around for a very long time.
> 
> I haven't a use for it, but various businesses use it if they don't
> leave PCs on 24/7.  Have the main server wake up by itself before the
> staff lob in, then progressively booting up each client PC with a
> minutes delay stops having a sudden mains power load, and staggers
> things that poll the server for everything under the sun.

The logic is clear. I just hadn't ever been aware of it, thanks.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 17:47 -0600, Roger Heflin wrote:
> I checked a mb bios from 10 years ago and it is in it.
> 
> And I checked a 2 year old one and both have it.
> 

Yes, I just verified that mine has it (and it's around 10 years old).

> RTC Alarm Power On
> Allow the system to be waked up by the real time clock alarm. Set it
> to By OS to let
> it be handled by your operating system.
> 
> web page for setting the timer from linux:
> https://www.linux.com/training-tutorials/wake-linux-rtc-alarm-clock/

That's very helpful, thanks.

I may play with it, however the factory reset on the plug seems to have
worked. 8am came and went without incident, so it looks like I can use
the plug as intended. However the BIOS option is there if it proves
unreliable.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 16:25 -0600, Roger Heflin wrote:
> No, we both mean use the bios timed wake up to wake up your machine
> instead of the plug for anything.

Ah, I misunderstood. I hadn't realised that was even possible.

For now, I've factory reset the plug and will see what happens tomorrow
morning. If the problem persists, I'll look into the BIOS option.

Thanks.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 14:13 -0800, Samuel Sieb wrote:
> On 11/16/23 04:03, Patrick O'Callaghan wrote:
> > I had installed this thing as a simple solution to waking up the
> > hibernating machine without having to futz around with Wake On Lan,
> > which I have no experience with and which apparently can be flaky
> > depending on what the BIOS supports. A more reliable device might
> > be
> > the answer.
> 
> Depending on the MCU inside and hardware accessibility, you might
> look 
> into something like https://esphome.io/ to replace the firmware.  I'm
> involved with that project and it's very nice for getting local
> control 
> of your own devices.

That would seem to be way outside my comfort zone/area of competence.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 19:09 +0100, Ralf Corsépius wrote:
> 
> 
> Am 16.11.23 um 14:39 schrieb Patrick O'Callaghan:
> > On Thu, 2023-11-16 at 07:08 -0600, Roger Heflin wrote:
> > > You might look in the bios.  Some of the bioses have an option to
> > > turn on at a specific time each day.
> > 
> > The plug has no accessible BIOS, if that's what you mean.
> 
> I guess, the OP meant your PC.
> 
> Some PCs have some kind of a "timed wake up" in their BIOS.

Sorry if I wasn't clear. I'd already discarded that explanation as I
could see the LED on the plug turning off and on again at the time,
i.e. it was the plug itself that was causing the problem.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 11:04 -0500, Go Canes wrote:
> On Thu, Nov 16, 2023 at 8:43 AM Patrick O'Callaghan
>  wrote:
> > I've emailed the manufacturer (TP-Link), though without much
> > expectation of a reply.
> > 
> > If it is as you say, it's a design fault. The power cycle is a
> > definite
> > Off, wait a second or so, then On.
> 
> I have TP-Link KASA EP10 smart plugs and I don't see this behavior,
> and at least one plug controls a device that would definitely be
> impacted by such a power cycle.
> 
This is a PM110, which is a low-end device. I'll take a look at the
EP10 as a possible alternative. Of course it's always possible that I
have a faulty one,

> For the sake of completeness you might double/triple/quadruple check
> that the plug doesn't have a schedule programmed.

I already did that.

>  Assuming it doesn't
> have its own schedule, is there any chance the plug is a member of a
> group/scene/etc such that while the *plug* may not have a schedule,
> it
> is impacted by something else that does?

There are no other smart plugs on the network. There are several Amazon
Echo devices, which I think also implement the Matter protocol and can
in theory talk to the TP-Link, but I've never configured them to do
this and they don't have any scheduled activities.

> Maybe try a factory reset on
> the plug (assuming there is such a reset)?

The docs say there is a factory reset, so I might try that.

poc
--
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 07:15 -0600, Roger Heflin wrote:
> Given it is the same time each day, I am guessing "feature" rather
> than bug.
> 
> It may also be that the power plug resets/reboots itself once a day
> and that the reboot does a quick power blip that is quick enough to
> not matter for non-computer devices.
> 
> Device makers love to find and implement half-assed "solutions" to
> problems.   I would guess they had some sort of memory leak (most
> software does) and will stop working after a week or 2, and instead
> of
> finding and fixing the leak chose to reboot it daily so that the leak
> is not an issue.   And on a reboot they don't have a way to maintain
> power during the reboot and decide a quick blip is ok.
> 

I've emailed the manufacturer (TP-Link), though without much
expectation of a reply.

If it is as you say, it's a design fault. The power cycle is a definite
Off, wait a second or so, then On.

poc
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 07:08 -0600, Roger Heflin wrote:
> You might look in the bios.  Some of the bioses have an option to
> turn on at a specific time each day.

The plug has no accessible BIOS, if that's what you mean.

poc
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 22:15 +1030, Tim via users wrote:
> On Thu, 2023-11-16 at 09:59 +0000, Patrick O'Callaghan wrote:
> > Seriously though, in the end I did watch it with hibernation
> > disabled,
> > and sure enough the "smart" power plug turned itself off and on
> > again
> > at 08:00.
> > 
> > I've checked and rechecked the settings on the thing, and it
> > positively, definitely, absolutely does *not* have a scheduled
> > power
> > cycle at that time. In fact I'd completely disabled it during the
> > experiment (but of course left it connected), yet it still decided
> > to
> > do this entirely of its own volition.
> > 
> > In other words, it's a bug in the plug.
> 
> I have a few of those things, thankfully none of mine have done that
> (using a widespread Tuya chipset).  I won't use them for anything
> critical, or risky.  Just a few light bulbs and a pedestal fan.  I'd
> never trust them on something that might be a fire risk (heaters), or
> requires stability (computing equipment).
> 
> I wish I could hack them to be controlled by Linux in my LAN, rather
> than some server on the internet which might decide they're wasting
> too
> much money supporting hardware you bought a few years back.  Since I
> do
> have a PC on 24/7 it's certainly feasible in my situation.  And other
> people have things like Raspberry Pies that they're happy to leave
> one
> running all the time.

I had installed this thing as a simple solution to waking up the
hibernating machine without having to futz around with Wake On Lan,
which I have no experience with and which apparently can be flaky
depending on what the BIOS supports. A more reliable device might be
the answer.

poc
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot

[Patrick O'Callaghan]

On Thu, 2023-11-16 at 15:46 +1030, Tim via users wrote:
> Tim:
> > > Strategic attack kittens?
> > > 
> > > But being serious, is the hardware clock set correctly?
> 
> Patrick O'Callaghan:
> > As reported by hwclock, yes
> > 
> 
> Is everything else fine at 8?  You're not getting a mains power dive
> from something else coming on via a timer?

No, it was the "smart" plug. See my reply to Samuel.

poc
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot

[Patrick O'Callaghan]

On Wed, 2023-11-15 at 21:53 -0600, Robert Nichols wrote:
> > On 11/15/23 17:10, Samuel Sieb wrote:
> > > > On 11/15/23 15:02, Patrick O'Callaghan wrote:
> > > > > > On Wed, 2023-11-15 at 14:20 -0800, Samuel Sieb wrote:
> > > > > > > > On 11/14/23 05:52, Patrick O'Callaghan wrote:
> > > > > > > > > > My system (both F38 and now F39) reboots itself
> > > > > > > > > > every morning
> > > > > > > > > > at
> > > > > > > > > > 8am,
> > > > > > > > > > yet nothing in the cron configuration is telling it
> > > > > > > > > > to do
> > > > > > > > > > that and
> > > > > > > > > > I
> > > > > > > > > > don't see anything obvious in the journal to cause
> > > > > > > > > > it.
> > > > > > > > > > 
> > > > > > > > > > How can I figure out what is triggering this? I
> > > > > > > > > > know the
> > > > > > > > > > description is
> > > > > > > > > > vague, but there it is.
> > > > > > > > 
> > > > > > > > As mentioned, that isn't a reboot.  It's a hard power-
> > > > > > > > off. 
> > > > > > > > Have you
> > > > > > > > tried being at your computer at that time to see what
> > > > > > > > it's
> > > > > > > > doing?
> > > > > > 
> > > > > > Once I've tried Barry's suggestion of turning off the
> > > > > > programmed
> > > > > > hibernation, that'll be the next thing. It's usually a bad
> > > > > > idea
> > > > > > to
> > > > > > change more that one variable at a time.
> > > > 
> > > > I don't see how watching your computer is changing a variable.
> > > > :-)
> > 
> > Indeed, I believe any quantum effects from being observed can be
> > disregarded here.
> > 
> > Or hey, maybe your computer has fantasies about becoming a quantum
> > computer!
> > 
> > But seriously, you might check whether there is some BIOS power
> > control option that is doing this. You could try changing the
> > hardware clock to/from UTC and see if that affects the time when
> > this
> > occurs.

Thanks Bob, but it turned out to be the F**king smart plug. See my
reply to Samuel.

poc
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Re: Strange daily reboot - SOLVED

[Patrick O'Callaghan]

On Wed, 2023-11-15 at 15:10 -0800, Samuel Sieb wrote:
> On 11/15/23 15:02, Patrick O'Callaghan wrote:
> > On Wed, 2023-11-15 at 14:20 -0800, Samuel Sieb wrote:
> > > On 11/14/23 05:52, Patrick O'Callaghan wrote:
> > > > My system (both F38 and now F39) reboots itself every morning
> > > > at
> > > > 8am,
> > > > yet nothing in the cron configuration is telling it to do that
> > > > and
> > > > I
> > > > don't see anything obvious in the journal to cause it.
> > > > 
> > > > How can I figure out what is triggering this? I know the
> > > > description is
> > > > vague, but there it is.
> > > 
> > > As mentioned, that isn't a reboot.  It's a hard power-off.  Have
> > > you
> > > tried being at your computer at that time to see what it's doing?
> > 
> > Once I've tried Barry's suggestion of turning off the programmed
> > hibernation, that'll be the next thing. It's usually a bad idea to
> > change more that one variable at a time.
> 
> I don't see how watching your computer is changing a variable. :-)

Heisenberg?

Seriously though, in the end I did watch it with hibernation disabled,
and sure enough the "smart" power plug turned itself off and on again
at 08:00.

I've checked and rechecked the settings on the thing, and it
positively, definitely, absolutely does *not* have a scheduled power
cycle at that time. In fact I'd completely disabled it during the
experiment (but of course left it connected), yet it still decided to
do this entirely of its own volition.

In other words, it's a bug in the plug.

Given that I actually do want to reboot (i.e. resume) at 08:00, I could
just hibernate as normal and let it do its thing, but I don't like
magic solutions. I may decide to replace it if I can get a refund (or
even if not - it was cheap).

I appreciate everyone's efforts to help with this, even if in the end
it had nothing to do with Fedora, or hibernation, or my computer.

Thanks guys.

poc
___
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/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue