You are not logged in.
Still, I can't understand why the user is asked about enabling/loading it only in expert install mode. This is definitely not software freedom "my way".
It's a limitation of the installer. We would love to have it ask about non-free in all situations, but we couldn't get it to do that. We do have other ideas about how to solve this problem, so don't count on it being this way permanently.
For now, you have to choose expert install and a mirror to get it your way. The alternative would be to leave the non-free firmware packages off the isos, and then those who need it would not be able to get it their way.
If you select one of the expert installs (graphical or not) AND select a mirror, then you will be asked if you want contrib and non-free in sources.list.
Without doing those two things, if your hardware requires non-free firmware, then sources.list will include contrib and non-free and the necessary packages will be installed.
The live isos have some non-free wireless firmware installed and a script to remove those packages. Only the main repo is enabled in sources.list.
Here's a brief description -
Make a directory to work in. (or make a spare partition and mount it).
Do a debootstrap install in that directory.
Chroot into that directory and run commands like you were running that system.
All those libraries will not be used by the host system, because they are in the wrong place. They will be used in the chroot system, because they are in the right place for that system. (chroot means 'change root' so it treats your work directory like the root of the filesystem.)
I'm pretty sure you can find debootstrap instructions somewhere on this forum.
You could install beowulf (or ceres) in a chroot and build packages there. Another possiblity is to use pbuilder, which if I understand correctly, builds packages in a chroot for you.
OK, sorry for the trouble. Here's what happened - I extended the expiration date back on Sept. 9, but I wasn't able to send the key to a keyserver due to network problems. I even used that key a couple weeks ago and didn't notice a problem, because the key is not expired on my system.
Key was uploaded to pgp.mit.edu a few minutes ago, and it show up on their website. It may take a little time to propagate to other keyservers. Meanwhile, here's the sha256sums directly from the file I uploaded to files.devuan.org back when the isos got uploaded. (May?) It matches what's on the download site.
76584ce7183993306af8b00edc8ffce3a1dc69b10f0a97e0a6302d49b0a63858 devuan_ascii_2.0.0_amd64_desktop-live.iso
8f535c235897303a5d55266858ffe9fe6c4d3e830f609c77c406f6e4aed0befa devuan_ascii_2.0.0_i386_desktop-live.isoYeah, my key expired. I wasn't paying attention. I'm not at home right now, so I can't fix it. Post the sha256sum of the file you downloaded, and someone here can verify it.
Make sure you say exactly which iso you have.
apt-cache policy sylpheed
sylpheed:
Installed: (none)
Candidate: 3.5.1-2+b1
Version table:
3.7.0-3~bpo9+1 100
100 http://deb.devuan.org/merged ascii-backports/main amd64 Packages
3.5.1-2+b1 500
500 http://deb.devuan.org/merged ascii/main amd64 Packagesapt-get -t ascii-backports install sylpheedWhere is 3.6.99 coming from? Ascii backports has 3.7.0-3~bpo9+1. I don't know if sylpheed is like firefox and thunderbird that can just be downloaded from mozilla and unpacked (or automatically updated from upstream), or if it really needs to be installed by the package manager. I'd go for the backports version, if anything.
Edit: Actually, I'd probably just stick to whatever is in the repo unless I needed some feature in the backports version. I have a vague memory of disabling that update message in sylpheed, but it's been a few years since I used it.
When I 'right click' on a link, instead of getting the drop down menu so as to be able to select what to do ie: open link in new tab/window/private window, etc. it directly opens the link in a new tab.
To make things worse, this does not happen 'all' the time, it happens at totally random times and is also not related to the site I am viewing so I have been unable to reproduce it at will.
I get this occasionally. Usually, it happens to me in thunar, but it has happened in firefox. Usually, the first item in the drop-down menu is the one that gets selected, but a few days ago, I right-clicked on a firefox tab to look at the context menu there, and it "automatically" selected "Move to new window" which is the fourth item in the list.
I think the problem is caused by movement of the mouse when right-clicking. Normally, if you right-click and release, the context menu stays open. If you right click and drag the mouse to highlight an item in the menu and then release, that item gets chosen/executed. I'm pretty sure I was not careful with the mouse when I clicked on that ff tab, and I moved the mouse enough during the click to get that fourth item.
Mounting removable drives: I needed to add user to plugdev group. User was only in groups user and netdev. It works fine now.
Iso would not boot in virtualbox in bios mode, only in uefi mode. Isolinux menu has 'union=aufs'. It boots with that removed.
I committed the changes to the installer, but I don't know when the packages will be built.
https://git.devuan.org/devuan-packages/ … aller-base
https://git.devuan.org/devuan-packages/ … taller-gui
Did the error message tell you to run apt --fix-broken install? What happened when you did that? Were any other packages named? There must be more information in output.
Link to the package: https://sourceforge.net/projects/refrac … -2.3.6.deb
Welcome back, dzz! Good to see you in action again. I hope all is well.
New exegnu boots and installs fine. Boot screen says 'Debian'. Installing base-files might change that to Devuan. Putting something in /etc/lsb-release will also work.
Removable media show up on the desktop, but won't mount in konqueror. Error message:
No supported mounting methods were detected on your system.No boot splash on uefi boot in the installed system. (That one is my fault - an extra "/" in refractasnapshot.)
Refractainstaller did not disable autologin. Log shows an error for
sed -e expression #2 char #3: unterminated s command.
I looked at it, and it appears to be terminated correctly. Maybe the ^ on ^AutoLogin needs to be escaped. I'll try that.
All in all, the new exegnu looks good. I'm a little envious that you got it under 700mb. I couldn't do that with refracta this time.
Edit: Found it. There's an extra "sed -i -e" on line 1406 of refractainstaller-yad.
I just did this a few days ago. Thanks for blazing a trail. It seems to be getting easier. Here's what I did -
Installed Refracta 9 (ascii with xfce installed a la carte, lxdm, elogind)
Added beowulf main repo to sources.list. Did not disable ascii repos.
Pinned policykit-1 and its libraries to the ascii versions.
Then ran apt-get update && apt-get dist-upgrade
Installed the 4.18 kernel, ran apt-get autoremove (only 3 or 4 packages got removed)
Rebooted.
It seems to be working. Shutdown/reboot buttons work in desktop and login screen. Mount/unmount removable drives on the desktop works. And I just got out of a video conference in chromium, so camera and mic are working, too.
Here's the pin file I used: NOTE: It is no longer necessary to pin policykit packages.
Package: policykit-1
Pin: version 0.105-18*
Pin-Priority: 1001
Package: libpolkit-agent-1-0
Pin: version 0.105-18*
Pin-Priority: 1001
Package: libpolkit-backend-1-0
Pin: version 0.105-18*
Pin-Priority: 1001
Package: libpolkit-gobject-1-0
Pin: version 0.105-18*
Pin-Priority: 1001Please post your sources.list and any repos enabled under sources.list.d.
Also, post the output of apt-get install sshfs and dpkg -l *ssh*
What error message are you getting? I've never had a problem installing or using sshfs in debian or devuan.
What needed was setting it to none.
RESUME=none
Where did you set that?
Thanks! Added to my TODO notes. You just saved me from writing a bunch of code to make a live-usb from a snapshot work directory for cases where the iso is too big.
Or maybe i can use good sense (for a change) and use the Dell without worrying over the battery.
That's what I do on my 12-year-old Dell laptop. There always seems to be a wall outlet nearby, and I don't have to worry about working too long and running down the battery.
Good find! I'm copying the answer here so we don't lose it. We can use either of the first two methods. Third one is here just to be complete. (I'm taking his word that this works. I haven't tried it yet.)
All of this leads to the three ways to enable Ctrl + Alt + Backspace termination of the X server, at least on a systemd based system. First, as part of your X session startup you can run setxkbmap to specifically enable C-A-B, among any other XKB changes you're already making:
setxkbmap -option 'compose:rwin' -option 'ctrl:nocaps' -option 'terminate:ctrl_alt_bksp'
Second, you can manually create or edit a configuration file snippet in /etc/X11/xorg.conf.d or your equivalent to specify this. If you already have a 00-keyboard.conf or the equivalent, the option you want is:
Option "XkbOptions" "terminate:ctrl_alt_bksp"
(A trailing comma is okay, apparently.)
Third, if you have Fedora or perhaps any systemd-based distribution, you can configure this the official way by running localectl with a command like this:
localectl --no-convert set-x11-keymap us pc105+inet "" terminate:ctrl_alt_bksp
There is a bear trap lurking here. That innocent looking "" is very important, as covered in the Arch wiki page. As they write (with my emphasis):
You'll also need to install the linux-headers-<version> that matches your running kernel.
apt-get install linux-headers-$(uname -r)
I can confirm that with Xorg-7.7 installed (default settings) ctrl-alt-backspace also works.
Cheers,
A.
Huh. It's not working here, and I tried it on two different ascii installations, same version of xorg, with and without keyboard layouts plugin.
Here's one of the earliest mentions of the change I can find. I tried getting it working a couple times and failed, so I haven't tried it in years (until now.)
http://forums.debian.net/viewtopic.php? … ce#p429554
My first guess is that the uuid in the resume file in your initrd needs to be changed. Try running update-initramfs -u and see if that fixes it.
True/False, Left/Right. Sometimes they switch places on me when I'm not looking. Sorry about that.
I tried this in a VM today and it worked, but you should heed the advice of the others.
Presumably you mean after i've installed the OS from the .iso file referenced in some post above this one which i can't access at this point. And after installing the recommended .iso file in some partition and
rebooting to run the newly installed OS. At that point in the process i run the following commands to install the broadcom drivers. If that is correct, then 'uname' should in some way correspond to the version number or codename of the version that you recommend.
Yes, install the wireless firmware after you install the system to hard drive. I forget which iso you're using, too. If you're using one of the installer isos, you could use an ethernet cable during the installation and pull packages from the repository. The live-isos just install from the media.
uname -r is a command that returns the version number of the running kernel. Putting it in the package name like that is using command substitution, and you can run that apt-get command exactly the way I posted it. Bash will substitute the kernel version, and you'll get the right linux-headers package. It's also possible to represent it like this: linux-headers-`uname -r`. It does the same thing, and it's easier to type, but it's less visible in a forum post.