Description of problem: Update from 3.13.5-202 on 2.83GHz Intel Core2 Quad Q9550 CPU nouveau video driver on NVidia 9800GT Version-Release number of selected component (if applicable): 3.13.6.fc20.x86_64 How reproducible: Attempt boot Steps to Reproduce: 1. Pick from boot menu 2. 3. Actual results: Very slow progress bars stall out when all white Expected results: Booted system Additional info: Deleted it from start menu to keep running the older kernel.
If you remove 'rhgb' and 'quiet' from the command line, do you get more data as to what is stalling/crashing? Without some additional kind of data we aren't going to be able to make much progress here.
I am afraid I have a very similar problem. And it is difficult to provide information about it. I was running 3.12.9-201.fc19.x86_64, during which time two new kernels were installed. When I finally rebooted, neither of them would boot. They would not even get to the point where rsyslogd was started to provide any information of any sort in /var/log/messages, and they printed absolutely nothing on the screen despite my taking quiet and rhgb off of the grub command line for them. The failing kernels are all 3-13, with the most recent being 3.13.6-100.fc19. I assume this is basically the same as 3.13.6.fc20. I am also using an NVidia 9800GT with the nvidia proprietary driver as installed via kmod-nvidia-3.13.6-100.fc19.x86_64.x86_64 After choosing the kernel in grub, it just waits a while, then powers off and reboots. I don't know what other information I can provide. There is no information saved anywhere when it does not boot. Except... TWICE, of about 100 tries of various forms, it *did* boot. And one of these times was a time when I had the nvidia driver ready to go and it *did* use it and start up kdm, and then log me in, just fine. I then tried to reboot again and it hasn't done it since. But 3.12.9-201 works fine, over and over and over. (I had to try 3.12 many many times because I had to reboot to it to try making some change to try 3.13 again. But then I tried each of them quite a few more times anyway, just since it seemed weird that 3.13 did boot twice. But 3.12 never failed, so it does not seem likely to be hardware. But how could it be then that 3.13 did boot twice, but not a second time ever with the same configurations (I tried repeatedly)?) Would it be useful to see the *successful* /var/log/messages?
Created attachment 875374 [details] /var/log/messages for one of two successful boots with 3.13.6. All other attempts gave NO /var/log/messages All other attempts gave NO /var/log/messages, and NO messages on screen either. NOTHING.
This is very frustrating and disturbing. I have been trying to fix it for twelve hours now. I cannot understand why it booted a couple of times, and these other times it just sits there for up to maybe five seconds before it reboots but there are NO MESSAGES of any form. What is it doing for that time that means it is not putting out any messages? Maybe they are going to some other place, a different tty or something. I imagine it is some video problem, but I cannot tell which driver is used by who. I removed plymouth to get rid of one complicating factor, with no change. This is very frustrating and disturbing.
I installed Fedora 20 just to test the kernel there as well on a pristine system unpolluted by anything I might have done to my Fedora 19 install. Used the main default live image on DVD and installed from there. Same problem. It did boot the first time, but then after that only about one more times out of 10. Using the grub sub menu it installed (havent put it back yet to a grub on my Fedora 19, which I will), the Fedora 19 version of the kernel also booted about twice in 10. Again, I cannot provide any messages. I doubt upping the value in /proc/sys/kernel/printk will help much since there should be messages already and there are not. Pretty sure there are messages being generated, since it usually takes a while before it silently stops and reboots. It is like it is being confused about where to send them (where the console is). I tried various options, including headless, and could not get it to boot all the way, or to put out messages. Again, I can attach the log it generated when it did boot if that might be helpful. I hope eventually some new kernel works on my system or I will be stuck forever on 3.12 kernels...
Benson, please open a new bug for your issue. Note that we don't provide or support the proprietary nvidia driver, so make sure you have a fresh install in that bug report. Remove 'rhgb' and 'quiet' from your command line and see if you get further information. If not, install kernel-debug.
Setting needinfo again from the original reporter.
Okay I will thanks. May take a while as I am worn out from this and have to use my only machine for real things as well. The install I mentioned second was totally pristine, with the only commands I ever executed on it being "yum update kernel" and "reboot" (Fedora 20 comes with a very old kernel). I removed rhgb and quiet from my command line a long long time ago and tried many other things as well. Do you see anything fishy in the /var/log/messages for the successful boot that might explain anything? Maybe something acpi related?
This bug is being closed with INSUFFICIENT_DATA as there has not been a response in 2 weeks. If you are still experiencing this issue, please reopen and attach the relevant data from the latest kernel you are running and any data that might have been requested previously.