| Summary: | 6sta1 classic iso fails to boot from USB on UEFI system | ||
|---|---|---|---|
| Product: | Mageia | Reporter: | Len Lawrence <tarazed25> |
| Component: | Release (media or process) | Assignee: | Anne Nicolas <ennael1> |
| Status: | RESOLVED OLD | QA Contact: | |
| Severity: | critical | ||
| Priority: | release_blocker | CC: | cae, fri, lebarhon, marja11, sysadmin-bugs, tablackwell |
| Version: | Cauldron | Keywords: | 6sta1, 6sta1.5 |
| Target Milestone: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | NEEDINFO | ||
| Source RPM: | CVE: | ||
| Status comment: | |||
|
Description
Len Lawrence
2016-05-03 17:01:43 CEST
Marja Van Waes
2016-05-03 17:02:49 CEST
Keywords:
(none) =>
6sta1 Same thing here error: file '/isolinux/x86_64/vmlinuz' not found. alloc magic is broken at 0x78992d80: 788ddb00 Mageia-6-sta1-x86_64-DVD on real hardware CC:
(none) =>
lebarhon Not so for the sta1 classical x86_64 on DVD. Works fine on UEFI as far as I've taken it (to choosing a language for the install). CC:
(none) =>
tablackwell I can confirm. UEFI install from usb fails with above noted error. UEFI install from dvd does work and install ran without major issue and am able to boot to a working Plasma desktop. A couple small hiccups. script failed for lib64user Mandatory grub2-mageia-theme can not be installed (grub2-efi-2.02-0.git10270.6.mga6 is installed but the previous release grub2-mageia-theme-2.02-0.git10270.5.mga6 is contained on the iso) At Configuration stage when setting up wired network. Interface is still being set-up as eth0 which fails. Network can be properly set-up as enp8s0 after 1st boot. CC:
(none) =>
cae Len and Charles hit this bug when installing from USB Charles did _not_ hit it when installing from DVD @ André I suppose you tried to install from USB when you hit this bug? @ Tony Had you written the iso to a DVD when you did _not_ see this problem? tonyb: All these using DVD: try1: got to asking language to use then totally frozen. try2: failed very early when progress bar for loading program from DVD barely started: 'error opening symlink definitions file '/tmp/stage2/usr/share/symlinks'. try3: got to choosing language but screen instantly freezes at this point. Same as try1. try4: as per try2. Very flakey in how far it gets before crashing. Don't understand why the crash point varies so much. (In reply to Marja van Waes from comment #4) > @ André > > I suppose you tried to install from USB when you hit this bug? > Yes, indeed. Thanks for the fast replies, Tony & André @ Tony Your issues are about drakx-installer-stage2, could you please file two separate bug reports for them? Summary:
6sta1 classic iso fails to boot on UEFI system =>
6sta1 classic iso fails to boot from USB on UEFI system According to thunar the file isolinux/x86_64/vmlinuz is there, described as a DOS/Windows executable. Ownership lcl:lcl (my moniker) Permissions 400 ls gives 700. $ ls -l total 75588 -r--r--r-- 1 lcl lcl 45265972 May 20 14:06 all-nonfree.rdz -r--r--r-- 1 lcl lcl 27535064 May 20 14:06 all.rdz -r--r--r-- 1 lcl lcl 4599312 May 20 14:06 vmlinuz Correction: 444 Still valid on officially released sta1? Seem to work for me up to a point in installer partitioner where it wanted a bios something in /boot partition. I did not have time to read up on that so I installed old style, booting stick non UEFI instead... Beware isodumper may write bogus, Bug 18411 (in my case i ended up at grub2 prompt, but creating stick using dd, it worked) CC:
(none) =>
fri
Rémi Verschelde
2016-07-28 11:22:31 CEST
Priority:
Normal =>
release_blocker for me, just writing the usb with 'dorsync' works for UEFI. No problems booting it or with subsequent install. Does this problem still exist fr the OP? No problems now on UEFI or CSM installing single or multiple desktops and no problems with rebooting or nvidia.
Len Lawrence
2016-07-29 01:14:40 CEST
Keywords:
(none) =>
6RC One thing that I would like to strongly suggest to everyone writing to or installing from USB. Do both using USB 2.0 DO Not use USB 3.0 for either. From my very unscientific testing using USB 3.0 can cause read as well as a variety of other errors especially during the install. Any new tests on that one so that we can close this bug ? In reply to comment #14. I have been using USB3.0 consistently for a while now without any problems that cannot be traced to nvidia troubles and other things so for me the bug is no longer an issue. OK. I am going to say it is resolved. Status:
NEW =>
RESOLVED |