[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: non-starter HTTP installation (success, sort of...)



On 2014-03-18 03:48 (GMT-0400) Felix Miata composed:

Target: single core 32 bit 2.4GHz P4, 1G RAM, Matrox G400 AGP video.

Brain fart. Not Matrox, but rather the historically problematic onboard Intel 845G.

Using
    5925232 Mar 14 16:22 vmlinuz
274341888 Mar 17 11:59 squashfs.img
   38301760 Mar 17 12:02 initrd.img
from
http://alt.fedoraproject.org/pub/fedora/linux/development/rawhide/i386/

I tried to initialize an HTTP installation from mirrors.us.kernel.org via
Grub. Cmdline option inst.resolution=1024x768 was ignored (as far as it got,
far short of GUI startup), as were vga=791, video=1024x768 and net.ifnames=0.
Init progressed only as far as "target Initrd File Systems", about 40
seconds. Only keyboard response was to ESC toggling between bootup message
and a bar at bottom of screen that ended "Fedora 21", and CAD to reboot,
since a wait of more than 10 minutes produced no more activity.

Should it have proceeded?

I was able to install the next day using the next built vmlinuz/initrd/squashfs. Still there were several obstacles to a satisfactory process to completion:

1-http://fm.no-ip.com/Tmp/Linux/F/gx260/ contains /var/log/anaconda/'s contents. Actually used Grub cmdline stage2=hd:/dev/sda11:/squashfs.img repo=http://mirrors.us.kernel.org/fedora/development/rawhide/i386/os ip=192.168.0.26::192.168.0.50:255.255.255.0:gx260:p1p1:none nameserver=192.168.0.50 ipv6.disable=1 splash=verbose inst.usefbx vga=791 selinux=0 nofirew obviously is truncated from the actual Grub stanza content, which I actually shortened at runtime manually from the menu.lst content.

2-It took several failed attempts to learn it was pointless to try and allow the Intel driver to be used by Anaconda and add inst.usefbx to be able to proceed. Some attempts nothing except the mouse pointer ever painted. In others, some painting occurred, but it was obviously corrupt and unusable, likely a problem in the Intel driver's video buffer related to it defectively trying to share motherboard RAM with the kernel. (KDE is working fine on the installed system.)

3-Default route was not created automatically, something else I figured out via the several failed starts with failure to populate available software or repo config. I had to do route add default gw... myself on tty2. (While I was there I turned on the swap partition to avoid a possible memory shortage).

4-The partitioner is a horrible experience. I planned to set a / target of sda19 and mount nothing more until after booting the installed system. I eventually had to restart after a manual reformatting of the target sda19 before it appeared as a single item among the random ordered multiple copies of virtually every other partition among the multiple instances of "unknown". Total actual partition count is 35, so having multiple random ordered copies of most makes for a very long and difficult to navigate list. Exactly how many instances of "unknown" was impossible to count, because mouse response was glitchy and scrolling was hard to modulate. I'm guessing the list is actually sorted by UUID, which from a human eyes recognition perspective amounts to unsorted.

5-It wasn't clear to me whether vga=791 was ever obeyed. I attempted to start on a 21" CRT supporting up to 2048x1536. No matter what was on cmdline WRT attempts to set video parameters with the CRT connected (e.g. inst.resolution=1024x768; video=1024x768), the result was always everything impossibly small to use, what looked like 1600x1200 with 5pt or 6pt fonts most places except for headings. I eventually switched to a 17" 1280x1024 LCD, but still the cmdline video options were ignored leaving Anaconda and tty text uncomfortably tiny. Finally I was able to get by with a 15" 1024x768 LCD. It's really sad that the only apparent way to get legible installation text is to switch to a much smaller display incapable of any "high" resolution mode. This is not a new problem. It dates back to the new installer whenever that was, F18 or so.

6-net.ifnames=0 is/was ignored, leaving sole network connection as p1p1 instead of the desired eth0. Supposedly this is not a bug: https://bugzilla.redhat.com/show_bug.cgi?id=1029815
--
"The wise are known for their understanding, and pleasant
words are persuasive." Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]