Trying 17 with "testing" repo, no X

Forum Forums Official Releases antiX-17 “Heather Heyer” Trying 17 with "testing" repo, no X

This topic contains 14 replies, has 8 voices, and was last updated by eriefisher Apr 8-4:09 pm.

Viewing 15 posts - 1 through 15 (of 15 total)
  • Author
    Posts
  • #18334
    Member
    Avatar
    seaken64

    Hi all,

    So, I decided to try my hand at running an antiX-17 system with the “testing” repo. I’ve been learning a lot about repos and have a better understanding of how linux repos work to update and upgrade the system. But I’m still learning. I thought I could accelerate this learning curve by installing a testing repo system on another computer, while I keep the stable systems on their own machines.

    I installed antiX-17 Full 32-bit and chose the “testing” repo. I then did the updates and upgrade and dist-upgrade. After the “dist-upgrade” I do not get into X. I am dropped at a text mode login. I can log in and run inxi, lsmod, MC, etc. But I tried starting X but can get the X server to start.

    Is the command still “startx”? I used startx and this is what I get back:

    /etc/X11/xinit/xserverrc: 3: exec: /usr/bin/X: not found
    xinit: giving up
    xinit: unable to connect to the X server: connection refused
    xinit: server error

    I looked at xorg.0.log and dmesg and lsmod. The hardware looks good. I am getting all the good drm and glx stuff using the intel i915 driver.

    Maybe there is a problem with a missing file? Where do I look next to troubleshoot X server?

    Thanks for any help you can give to get me pointed in the right direction.

    Seaken64

    #18345
    Member
    Avatar
    skidoo

    I would not expect simply typing “startx” to work.
    Following a dist-upgrade operation, I would attempt a normal shutdown then reboot. (do this at least twice)
    After several reboots, if the problem persists, I might add nomodeset to the bootline…

    Afterward, if still faced with a tty, I would type “sudo init 5”.
    Faced with a tty prompt in runlevel 5, I would type “sudo slim” (and be prepared to copy any error messages).
    If still no joy, I would begin grepping /var/log/* for warning//error messages.

    #18348
    Forum Admin
    Dave
    Dave

    You can look in /var/log/ at the xorg log and the slim log. There should also be a log in your home folder for xorg if you start things manually. If you want to try startx… run like this
    startx /usr/local/bin/desktop-session rox-icewm
    To get a normal antix type session.

    Follow skidoo suggestions first. Also one more stupid sounding suggestion. When you turn the computer on just leave it for 5-10 minutes and see if the slim log in comes up. I have had it twice now after upgrading to testing in the last few days that this long wait makes it work. I have yet to figure out why other than slim is waiting for the x server to start excepting connections. I could get slim to come up sooner by doing service slim stop && service slim start but am not able to login. Running startx would not work (or rather would take 5min). And the very curious thing was trying lightdm as the login manager which comes up right away….

    Computers are like air conditioners. They work fine until you start opening Windows. ~Author Unknown

    #18349
    Forum Admin
    anticapitalista
    anticapitalista

    If using slim on Testing, make sure you have consolekit installed (and you probably need elogind as well)

    Philosophers have interpreted the world in many ways; the point is to change it.

    #18350
    Member
    caprea
    caprea

    Honestly, better make a whole new start.
    It’s not an old set up system where it’s worth saving something, it’s a new installation, isn’t it?
    Something went completely wrong.
    Do a new installation again, and take a closer look at what happens during the dist-upgrade.
    Follow the instructions d_o gave you in the mx forum, and did you also do an apt-get autoremove ?
    To get new stuff configured, old stuff has to be removed.
    If you’re unsure, you could even say no at the terminal first,and post the output here.
    Like d_o mentioned,the whole procedure has to be repeated until you see the normal terminal output ending.

    #18356
    Forum Admin
    anticapitalista
    anticapitalista

    If you do decide to do a new installation, I highly recommend you start with the latest antiX-17.3.1 iso (released 5 weeks ago) and not one released 15 months ago.

    Philosophers have interpreted the world in many ways; the point is to change it.

    #18363
    Forum Admin
    dolphin_oracle
    dolphin_oracle

    as I’ve mentioned elsewhere, the update to testing doesn’t complete with one dist-upgrade. You’ll likely need to do a couple for the update to complete

    #18365
    Member
    fungalnet
    fungalnet

    But startx worked before you shifted to testing?
    It might be that you were asked somewhere to keep your previous .conf file or install the distribution’s version and you may have kept something incompatible. What if you try to reinstall xorg xorg-xinit your desktop or window manager.
    Do you have a .xinitrc file in your home directory?
    What about xf86-video-intel

    • This reply was modified 5 months, 2 weeks ago by fungalnet.
    #18374
    Member
    Avatar
    skidoo

    Do you have a .xinitrc file in your home directory?

    In a recent discussion, antiX-Dave mentioned that ~/.xinitrc is not sourced during normal antiX operation (telinit 5)

    #18449
    Member
    Avatar
    seaken64

    Thanks everyone! You’ve given me a lot of good tips. I haven’t had the time to work on this project this week. I will get back to it this weekend and try out your tips and get back to you.

    Seaken64

    #18458
    Member
    caprea
    caprea

    My experience
    I know that some people like to avoid the sudo apt-get autoremove, I always do the apt-get autoremove, if the terminal tells me so.
    It’s maybe possible to avoid this if you dist-upgrade your normal stable installation.
    But it is surely necessary, if you eg. dist-upgrade from jessie to stretch, or from stretch to testing-repos.
    Recently I did set a new installation of antix17.3.1 to testing.
    Even locale couldn’t get configured without sudo apt-get autoremove.
    In principle, it was
    sudo apt-get update
    sudo apt-get dist-upgrade
    sudo apt-get autoremove
    sudo apt-get dist-upgrade
    sudo apt-get autoremove
    sudo apt-get dist-upgrade

    Nothing special happened here except grub was updated, there’s a video somewhere of d_o , how to do this (if you have any questions about that)

    A completely different case is when the terminsal says : the following packages will be removed
    That’s dangerous(maybe)Then you always have to watch : What is happening there ?

    You also will get asked very often what version of configuration files you want to keep, during the procedure, I always took the new ones on this fresh installation.

    Good luck ! (Remember, these first whole dist-upgrade procedures are the most difficult part, after that it is actually quite easy to handle a testing.IMO.In the end it remains what it is called:testing)

    #18468
    Member
    Avatar
    seaken64

    Thanks again to everyone for the help.

    I decided to take the advice to re-install, and I used the latest 17.3.1. I deleted the old install and reformatted and I chose to check for bad blocks with the installer.

    After the install completed I ran:
    sudo apt-get update
    sudo apt-get dist-upgrade

    After this first dist-upgrade finished I entered:
    sudo apt autoremove

    After hitting enter there was a long list of packages to be removed and then apt started doing it’s thing. After a short time I noticed that it seemed to be doing another dist-upgrade. I did not enter this command again, it just started doing it on it’s own. Maybe it was on the screen with a Y/N? I’m not sure. Anyway, it was clearly now running dist-upgrade again.

    After the dist-upgrade finished I entered dist-upgrade again and it returned with 0 upgrades available. It was finished.

    I rebooted, and this time I got into X with the IceWM desktop. I reset the screen resolution by making an xrandr entry in the IceWM startup file. I used Package Installer to install SeaMonkey, Pale Moon, Chromium, and SMPlayer. SMTube and streamtuner2 were already the latest versions. I installed Remmina using the apt-get commands. I tested them all and rebooted again. All is good.

    So, now I wonder when I might run into some kind of problem. Everything feels like a regular antiX install so far. I guess time will tell.

    Thank you all again for pointing me in the right direction.

    Seaken64

    #18469
    Member
    Avatar
    seaken64

    The differences between this successful install and the failed install:

    The failed install was using 17.3.
    I did not check for bad blocks when formatting.
    When I did the upgrade I ran it like this:
    sudo apt-get update
    sudo apt-get upgrade
    sudo apt-get dist-upgrade
    I did not run apt autoremove until the end
    I changed the screen resolution with an xrandr command in the startup file before I ran the upgrades.

    The successful install used 17.3.1
    I selected for the install to check for bad blocks when formatting
    I did the upgrade like this:
    sudo apt-get update
    sudo apt-get dist-upgrade
    sudo apt autoremove
    followed by another dist-upgrade (even though I don’t think I actually entered that command. I seems to have been automatically run by the apt autoremove command).
    I did not change the screen resolution with xrandr command in the startup file until after all upgrades were done.

    I don’t know if any of this is significant. All I know is that the second try, using 17.3.1, worked.

    Seaken64

    #20223
    Member
    Avatar
    seaken64

    I haven’t been able to do a lot of “using” with this test system. But I have been firing it up every few days and doing an update and dist-upgrade. A few weeks ago the desktop got messed up a bit – some missing icons on the panels and no theme colors. I left it for awhile thinking I would get back to it. Today I opened it up and the desktop wallpaper was missing, only the area behind the conky had a portion of the image.

    I ran another update and dist-upgrade and rebooted. Now everything is back to normal.

    Is this usual for the “testing” branch? If something gets messed up a bit I can wait it out and see if the next upgrade fixes it? Or should I be trying to figure out why the image was missing and the theme was messed up? I’m still a rookie so I wasn’t sure how successful I will be in troubleshooting.

    I haven’t made any real changes yet. Just the default install so far. Maybe if I had been tweaking and changing the look and feel I should be working hard to see if I can fix it myself. But for now I’m not even sure where this stuff is configured.

    Seaken64

    #20270
    Member
    eriefisher
    eriefisher

    A while back I wasn’t paying attention and was unaware of the eudev switch over and X was totally remove due to the compatibility issues. It took a while but I finally found out why and slowly got everything up and running. I had to install everything one at a time until I found what the problem was. The option was to move to my wife’s win10 box and search for issues but that wasn’t happening. After I got it all up and running again I decided to reinstall anyway with a later version. The one I was using had been running for a long time so I treated myself to a fresh install. I’m glad I did.

    eriefisher

Viewing 15 posts - 1 through 15 (of 15 total)

You must be logged in to reply to this topic.