You are not logged in.
shwsh, did you do this on jessie, ascii or ceres? 3.24.3-1 is in buster/sid.
What you find depends on how you look. I just added this line to my sources:
deb http://auto.mirror.devuan.org/merged ascii-proposed-updates main contrib non-freeAnd now I see this - chromium-60 is in ascii-proposed-updates
# apt-cache policy chromium
chromium:
Installed: 57.0.2987.98-1~deb8u1
Candidate: 57.0.2987.98-1~deb8u1
Version table:
60.0.3112.78-1~deb9u1 0
100 http://auto.mirror.devuan.org/merged/ ascii-proposed-updates/main amd64 Packages
59.0.3071.86-1 0
100 http://us.mirror.devuan.org/merged/ ascii/main amd64 Packages
*** 57.0.2987.98-1~deb8u1 0
500 http://us.mirror.devuan.org/merged/ jessie/main amd64 Packages
500 http://auto.mirror.devuan.org/merged/ jessie-security/main amd64 Packages
100 /var/lib/dpkg/statusYou posted the following. There's no error message saying that stretch-backports does not exist, so you must have had stretch-backports in your sources at some point:
#apt-get -t stretch-backports install chromium
Reading package lists...
Building dependency tree...
Reading state information...
chromium is already the newest version.
0 upgraded, 0 newly installed, 0 to remove and 38 not upgraded.
To see if you pulled in packages from backports or testing, you can run
aptitude search ~i -F"%p# %v# %t#"but I'm not sure it will be entirely accurate. I'm running jessie, and it shows most of my packages are from "stable" with a few from "jessie-backports". If there are packages from ascii or stretch, they should show up as "testing".
For packages that have not been devuanized, you won't be able to tell whether you pulled it from debian or devuan unless you can find a version mismatch, Where debian has a different version than devuan, and you have installed the version from debian. Right now, the only example I can think of is chromium-60 in stretch security vs. chromium-59 from ascii. If you were running ascii and had chromium-60 running and you didn't get it from the ceres repo, you would have gotten it from debian.
For reference, here are some other useful commands: https://dev1galaxy.org/viewtopic.php?id=511
leloft,
I'm sure Dan meant sources.list and sources.list.d. I don't see any debian sources in what you posted. What packages did you get from debian that you should not have gotten?
I can offer half an answer to my own question (Q2, post#6):
If Amprolla is down or otherwise unavailable, apt-get appears to use the underlying debian repos in consequence. This results in a whole bunch of unauthenticated packages (because I have the devuan keyring not the debian) including packages which are normally held back. Although this constitutes using mixed repos, it appears like normal behaviour to apt-get, and so it simply gets logged as a striaghtforward upgrade. This has happened three times now: it appears that this behaviour is reproducible. I don't know enough to call it a bug, but it seems serious enough to warrant flagging up. Perhaps someone who knows more than me could confirm and escalate if necessary. For the rest of us noobs, just exercise caution if Amprolla is unavailable.
I checked with someone who knows more than both of us put together (CenturionDan):
if that happens then there is a debian stanza in either /etc/apt/sources or /etc/apt/sources.d/
Maybe this will help:
https://dev1galaxy.org/viewtopic.php?id=1571
Until you get more space on /dev/sda3, nothing you do can be trusted to work. Get it to less than 95% full. And figure out what's taking up all the space in /var/log. (Please don't answer with just "logs". Which ones and how big are they?)
To post text from console:
cat <filename> | nc termbin.com 9999and then post the link it gives you.
Use wget to download the iso. You'll need to remove the /download from the end of the file's url. If you get disconnected before the download is finished, run 'wget -c <url>' and the download will pick up where it left off.
sha256sum files would be helpful.
(Also a good idea for miyo to check the sha256sums on sf after uploading the files. Need to log in through ssh to do that.)
It means I'm confused and thinking of playing media files with a plain ssh connection instead of using sshfs. The latter works correctly and plays on the local machine. Thanks.
You won't find any eudev files. It has to lie to everyone and pretend to be udev, so a lot of the udev files you find are really the eudev files in disguise. And all that /lib/systemd stuff is for udev/eudev.
If you just want to access the server from the local network, nfs will work. You can either add the nfs shares to the fstabs of the clients, or you could install autofs and set it up so the shares get mounted automatically when someone tries to access those files. There's also sshfs, which is more secure, but won't work right for audio/video (it plays on the server instead of the client.)
Look for a howto on setting up nfs on debian. Same for autofs if you want that.
You can highlight the text you want in a code box and then click on the code link above the message box.
Yours:
2386824 ./logMine:
3984 /var/logHow big are your log files? Mine are all measured in kb. If yours are in the mb range, then maybe logrotate isn't working right. It's safe to remove all the archived logs (the ones that end with .gz). You can also get rid of any packages saved from upgrades (in /var/cache/apt/archives) by running
apt-get autoclean(gets rid of the ones that are obsolete) or
apt-get clean(gets rid of all of them).
I just did an upgrade to ascii, and I was having the opposite problem. Synaptic and gparted would start without asking for a password. Only worked with the primary user. I noticed that policykit-1-gnome had been removed, so I reinstalled it, and now synaptic and gparted won't start for the user. User can still use pkexec to create another user but has to use the full path for the adduser command.
In the upgraded Refracta:
Tried replacing mate-polkit with lxpolkit. Didn't help. Tried adding policykit-1-gnome (with each of the other polkits) didn't help.
In the upgraded Devuan:
Tried adding policykit-1-gnome first, and now gparted and synaptic don't start. Tried from the menu and from the terminal with pkexec.
However, I can do 'pkexec /usr/sbin/adduser buford' and it works. Had to use the full path with adduser.
I ran into this a few days ago and ended up downloading it from https://get.adobe.com/flashplayer/otherversions/ for a friend. We can't get it to work. I already have flashplugin-nonfree installed from long ago, and when I go to about:plugins, I don't see it listed. And neither of us could see some items on one site that requires flash. Debian wiki looked helpful, but as I said, the plugin didn't seem to work and doesn't show up in the list of plugins. I don't know what's going on with it.
Using xfce here. This was an install from the desktop-live iso. It's actually a mixed ascii/ceres now. I tested it first in pure ascii, then added ceres and did a dist-upgrade without removing the ascii sources. I had no problems during the upgrade. And pkexec acts the same in both.
I did notice in another installation that lxqt did it right. I guess the lx teams know what they're doing.
Same behavior in a devuan jessie that got upgraded to ascii, and then again when upgraded to ceres.
pkexec allows primary user to install and remove software (didn't test gparted this time)
It does not allow primary user to create a new user, and it does not let the second user install software.
I will get around to filing a bug report.
It's likely that eudev will be configured,,,
Future tense. It actually might already be in the latest build of eudev, but I'm not sure if that's in the repo yet. Need to check parazyd's test repo.
The live-boot stanza is correct.
Easy fix for the change in network names - revert to the old names by adding "net.ifnames=0" to the boot command.
You can add it to /etc/default/grub, next to "quiet" (will only affect the default boot choice) or in the line below (will affect all boot entries.)
Before making a snapshot, add it to /usr/lib/refractasnapshot/iso/isolinux/live.cfg (ang grub.cfg.template if you're enabling uefi boot.)
Note: this will be a config option in a future version of snapshot.
Note2: It's likely that eudev will be configured to reverse this behavior (i.e. you'll need net.ifnames=1 if you want the new names.)
As I mentioned in another thread, you need to tell apt to install linux-base from backports when you install the bpo kernel.
apt-cache policy linux-base
linux-base:
Installed: 3.5
Candidate: 3.5
Version table:
4.5 0
100 http://us.mirror.devuan.org/merged/ ascii/main amd64 Packages
100 http://auto.mirror.devuan.org/merged/ ceres/main amd64 Packages
4.3~bpo8+1 0
200 http://packages.devuan.org/merged/ jessie-backports/main amd64 Packages
*** 3.5 0
500 http://debian.csail.mit.edu/debian/ jessie/main amd64 Packages
100 /var/lib/dpkg/statusOn to the main topic. (I haven't done this in a long time, so you're getting the quick version.)
Install both kernels in the host system.
Make a snapshot (with save_work="yes")
Whichever kernel is symlinked with /vmlinuz will be in the copy. (same with initrd) That will likely be the backports kernel.
Copy the other kernel and initrd into place. Something like
cp /boot/vmlinuz-3.16-blah /home/work/iso/live/vmlinuz1
cp /boot/initrd.img-3.16-blah /home/work/iso/live/initrd1Edit your boot menus to add stanzas for vmlinuz1 and initrd1.
Re-run refractasnapshot and choose the 4th option (re-run xorriso, no-copy, no-squash)
Potential problem (that you already thought of): 4.9 kernel needs newer live-boot and live-config than the one used by 3.16. The older version of live-* uses aufs and the newer uses overlay. The reason the newer kernel needs the newer live-* is because aufs is not in the kernel.
I don't know if 3.16 will work with newer live-* packages. If it does, you'll only have to do this once. (in theory)
Other option would be to recompile the 4.9 kernel and add aufs support. Then you could use the older live-* packages.
Oh... just thought of another way to do it.
Make an iso with 3.16, put it on a usb (using refracta2usb) and create two persistent loopback files (with full persistence). Make them big so you can install packages.
Keep one persistent volume for running 3.16.
Boot the other persistent volume and add the backports kernel and live-* packages.
Go back to the host system, mount the usb, mount the loopback file with the bpo kernel and copy that vmlinuz and initrd into the /live dir as described above.
Then dd the usb stick into a file (named something.img). File size might be an issue in regards to upload and download.
MiyoLinux wrote:Kernel 4.9 is available via the Backports repository.
Negative unfortunately, I can't get it to install anymore, it's asking for a newer version of linux-base which is not available in backports or jessie main.
It's there in backports. You have to tell it to install linux-base when you tell it to install the kernel. Use
'apt-get -t jessie-backports install linux-base linux-image-4.9-whatever'
vagrant file is here -
https://files.devuan.org/devuan_jessie/virtual/
Refracta2usb is now included on all Vuu-do isos, if anyone can figure out how to work it please let me know. ;-)
I can! I can!
1. Make sure automount is NOT enabled for usb.
2. Plug in your thumb drive.
3. Run 'refracta2usb' from a root terminal (or sudo).
4. Click on the Help option. Read the first few paragraphs.
Keep in mind that it's like a Swiss army knife - it has a bunch of utensils in case you ever need them, but you'll probably never use all of them. Just pick the one(s) you need at any given time.
Hint: create a vfat that's big enough to fit a few isos and an ext partition for other uses, but don't use it yet. Get one iso running then talk to me about persistence (or read the rest of the Help).
If you get a popup error window and you see the word "rescan" in it, go back to the main menu and Rescan. (It happens a lot if you do multiple procedures in the same run. It's not a bug. That's just the way it is - stick gets unmounted after you do a task.)
@greenjeans - newer live-boot and live-config (not live-build) are needed with 4.9 kernel due to changes from aufs to overlay. Start a separate thread about multiple kernels. I've done it in the past. Not sure if it's possible in this case.
Warning: The home dir /run/uuidd you specified can't be accessed: No such file or directory
Were you installing software in a chroot? That would account for /run being absent.
The error about missing /home/devuan is weird. I don't know what to say about that.
save_work=yesKeep it that way unless you need to reclaim the space. (You already said you don't need to.) It will save time because rsync only copies changes after the first time.
Network problem: easily solved by adding 'net.ifnames=0' to the boot command. Since you saved your work, you can very easily add this in without running the whole process again. To test it, press TAB at boot menu and edit the line.
- Edit /home/work/iso/isolinux/live.cfg (the boot menu)
- Run refractasnapshot again and choose the 4th option (Re-run xorriso, (make iso, no-copy, no squash). It'll take a minute or so.
(Note: this will only work for changes in the root of the CD. It won't work for changes in the copied filesystem.)
Next time you do a full run, start it from terminal in debug mode -
refractasnapshot -d
or
refractasnapshot-gui -d