Description of problem: There seem to be numerous problems particular to a machine with an Intel i9 Decacore processor and nvidia GTX1080Ti graphics connected to a 3K monitor via HDMI. Booting from a USB stick results in the process hanging without any text or graphics displayed. Sometimes the boot menu appears and selections can be made but more often the menu is obscured by an empty black rectangle like the background of the edit screen. No virtual consoles accessible when it is hanging. Discovered by accident with one of the tests that pulling out and pluggin the USB stick kicked something into life; the "Which language?" query appeared and after the selection the screen switched to console mode and the process hangs again. There is information on the screen - picking out a few lines: nouveau 0000:65:00.0 secboot: error during falcon reset: -110 ....gr: init failed, -110 [...] vboxsf: error -19 connecting to guest PCI-device. [...] ACPI BIOS Error (bug)..... [...] EDAC skx: ECC is disabled on imc 0 There is a mouse cursor on screen. On tty2 login fails for live -> login incorrect Version-Release number of selected component (if applicable): Mageia7 beta1 Plasma/GNOME/Xfce Live iso - second round How reproducible: Failure to enable graphics - always Corrupted menu display - intermittent Steps to Reproduce: 1. Dump Live iso to USB stick 2. Use BIOS to boot it on the specific, or similar hardware 3. Attempt to select a menu item or just hit Return for the first entry 4. Wait for the process to hang without any screen output 5. Replugging the USB drive during the hangup is optional
The title of this bug may not be a correct description of this problem. The "because" is actually unknown.
Hi Len, Can you mount the USB stick you installed to from a working system and then: * find the directory with a very long hexadecimal name in /that_USB_stick's_root/partition's/var/log/journal/ * then run, as root, journalctl -D /path/into/that/hexadecimal/directory/ > log.txt * then attach log.txt to this bug report. (Compress it with "xz -9 --text log.txt" if it's too large to attach.)
CC: (none) => marja11Assignee: bugsquad => isobuildKeywords: (none) => NEEDINFO
Keywords: (none) => 7beta1
Thanks Marja. Shall try that later today, if possible.
Right - got it. Booted Live Plasma on an Intel i7 with nvidia GTX970. Copied the compressed file to another USB pendrive.
Created attachment 10503 [details] /var/log/journal/ journal file from Live Plasma session
Reassigning as this looks to be a driver problem.
Assignee: isobuild => kernelKeywords: NEEDINFO => (none)CC: (none) => mageia
This looks to be a very specific problem with the particular hardware setup used here. What happens is that the monitor goes dead very quickly if there is no signal. It happens during the reboot process most times. The login prompt is there but the monitor does not seem to accept input once it goes dead. One or several sequences of switch-off/switch-on succeed in waking it up or sometimes a Ctrl-Alt-Backsp to restart the X server. It is a nuisance and it is difficult to figure out what to do about it or which driver might be involved. I have not yet tried a different monitor. This one is running on an HDMI connection.
With reference to comment 7, swapped the monitors between two machines, including the hdmi and dvi connections. Thereafter there appeared to be no more problems. This is difficult to quantify so the bug should be regarded as fixed I think, unconfirmed anyway.
Status: NEW => UNCONFIRMEDEver confirmed: 1 => 0
I was unsure whether the problem has 'gone away', or is still extant if you configure things in a certain way (monitor & connection). If that is the case, it remains valid to leave open. OTOH If you are confident that the problem is dead, please do close the bug. I use INVALID for little things that no longer matter.
We will assume the problem was not reproducible, or has been fixed in one of the updates we have released for the reporter's distribution. Users who have experienced this problem are encouraged to upgrade to the latest update of our distribution, and if this issue turns out to still be reproducible in the latest update, please reopen this bug with additional information. Closing as OLD.
Resolution: (none) => OLDStatus: UNCONFIRMED => RESOLVED